Home Docs Claude Developer Platform

One capture · api

One read of Claude Developer Platformapi-20261030T212610Z

396 pages moved out of 826 read.

Pages moved 396 significant first

Pages read 826 in this capture

Captured 30 Oct 2026 21:26 UTC

Corpus hash 68200c8a9acb corpus-hash

What this read moved

1-25 of 396, page 1 of 16

This capture is too large to show at once. Changes 1-25 of 396 are below, significant first; the rest are on the following screens.

about-claude/models/optimizing-for-cost-and-intelligence Changed · +101 / -101 lines

from line 57

5757

5858**What good looks like.**

5959

60* Over a full day of real traffic, agent loops read a median 84% of their input from the cache, and the top 10% of them, coding or not, read 94% or more[17](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

60* Over a full day of real traffic, agent loops read a median 84% of their input from the cache, and the top 10% of them, coding or not, read 94% or more[17](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

6161* Deep in a task, a well-built loop pays full price on under 1% of its input.

6262* Below about 80%, look for something breaking the cache (see [What breaks the cache](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#what-breaks-the-cache)).

6363

64Anthropic priced the DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) runs with and without the cache discount:

64Anthropic priced the DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) runs with and without the cache discount:

6565

6666<Frame>

6767 ![Chart, DeepResearch Bench II: caching cuts Fable 5.1 from $40.58 USD to $5.64 USD a task; Sonnet 5.5, $5.13 USD to $1.05 USD](https://platform.claude.com/docs/images/cost-intel/caching-repriced.svg)

from line 71

7171

7272Pricing other runs’ recorded token counts the same way, each model at its default effort, gives the average token cost per task without caching:

7373

74| Benchmark | Model | Without caching |

75| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------- | ------------------------------- |

76| SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Opus 5.5 | 3.3 times the cost with caching |

77| SWE-bench Pro | Claude Sonnet 5.5 | 3.1 times the cost with caching |

78| Claude Managed Agents task set[25](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), whose sessions are short | Claude Haiku 5.5 | 1.7 times the cost with caching |

79| A small agent that triages bug reports ([Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens))[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Sonnet 5.5 | 7.3 times the cost with caching |

74| Benchmark | Model | Without caching |

75| ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------- | ------------------------------- |

76| SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Opus 5.5 | 3.3 times the cost with caching |

77| SWE-bench Pro | Claude Sonnet 5.5 | 3.1 times the cost with caching |

78| Claude Managed Agents task set[25](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), whose sessions are short | Claude Haiku 5.5 | 1.7 times the cost with caching |

79| A small agent that triages bug reports ([Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens))[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Sonnet 5.5 | 7.3 times the cost with caching |

8080

8181The saving varies with how many turns a task takes, because shorter loops re-read less.

8282

from line 83

8383<Note>

8484 With Claude Haiku 5.5 on the Messages API, [compact a long conversation](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#manage-the-context-lifecycle) between tasks, before its prompt passes 100,000 tokens. Past that, every token of the request costs 5 times as much, output included. Cache reads and writes count toward the 100,000 (see [Long context pricing](https://platform.claude.com/docs/en/about-claude/pricing#long-context-pricing)).

8585

86 If you use [compaction at a token threshold](https://platform.claude.com/docs/en/build-with-claude/compaction-threshold#parameters) instead, set its `trigger` a little below 100,000 minus your prompt’s largest growth between requests (the default is 150,000). In Anthropic’s test with 10,000-token turns, a trigger of 90,000 cost 70% less per request than the default[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

86 If you use [compaction at a token threshold](https://platform.claude.com/docs/en/build-with-claude/compaction-threshold#parameters) instead, set its `trigger` a little below 100,000 minus your prompt’s largest growth between requests (the default is 150,000). In Anthropic’s test with 10,000-token turns, a trigger of 90,000 cost 70% less per request than the default[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

8787</Note>

8888

8989#### Pick the cache duration

from line 92

9292

9393The 1-hour duration costs more to write (2x the input price instead of 1.25x). A miss on either duration bills the whole prefix at the write price instead of the read price.

9494

95Choose the cache duration when a conversation starts. In Anthropic’s test, the first turn added at 1 hour to a conversation cached at 5 minutes wrote the whole prompt again, once: $0.80 USD at 100,000 tokens on Claude Opus 5.5, compared with $0.02 USD of input for a turn that read the cache[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

95Choose the cache duration when a conversation starts. In Anthropic’s test, the first turn added at 1 hour to a conversation cached at 5 minutes wrote the whole prompt again, once: $0.80 USD at 100,000 tokens on Claude Opus 5.5, compared with $0.02 USD of input for a turn that read the cache[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

9696

97Anthropic measured a small agent that triages bug reports ([Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens)) with pauses inserted before some turns to simulate a person’s delay[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). Accuracy stayed within run-to-run noise at every setting measured. Anthropic also measured extra requests that keep the 5-minute cache warm (keep-alive requests).

97Anthropic measured a small agent that triages bug reports ([Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens)) with pauses inserted before some turns to simulate a person’s delay[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). Accuracy stayed within run-to-run noise at every setting measured. Anthropic also measured extra requests that keep the 5-minute cache warm (keep-alive requests).

9898

9999To decide, count the gaps between consecutive requests in a conversation, and find your row in the following table:

100100

from line 125

125125

126126* **Claude Fable 5.1**, whose [cache reads](https://platform.claude.com/docs/en/about-claude/pricing#prompt-caching) cost 0.025x the input price.

127127

128 * **Pauses run for minutes.** On Claude Fable 5.1, keep-alive requests[19](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) cost 13% to 20% less than the 1-hour cache, with 10% of turns paused and with a 6-minute pause before every turn.

128 * **Pauses run for minutes.** On Claude Fable 5.1, keep-alive requests[19](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) cost 13% to 20% less than the 1-hour cache, with 10% of turns paused and with a 6-minute pause before every turn.

129129 * **Pauses run toward an hour.** On Claude Fable 5.1, only with a 45-minute pause before every turn did the 1-hour cache win, by about $0.12 USD on a session of 5 bug reports.

130130

131131The following chart plots cost per session against the share of paused turns on Claude Sonnet 5.5:

from line 152

1521522. Re-send the request’s headers with its body. Without the original request’s `anthropic-beta` header, the beta-gated fields in the replayed body are rejected.

1531533. Send it within 4 minutes of the previous request’s start, and every 4 minutes after that, until the next turn arrives or the session ends. Count from the request’s start, because the [cache’s 5-minute lifetime](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#how-prompt-caching-works) runs from there.

154154

155That is the [pre-warming request](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#pre-warming-the-cache): it refreshes the cache’s lifetime, returns no output, and costs little more than the cache read. Once the cache has expired, the next request, keep-alive or not, re-writes the prefix at the cache write price[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

155That is the [pre-warming request](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#pre-warming-the-cache): it refreshes the cache’s lifetime, returns no output, and costs little more than the cache read. Once the cache has expired, the next request, keep-alive or not, re-writes the prefix at the cache write price[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

156156

157157A `max_tokens: 0` request is rejected in the following cases:

158158

from line 190

190190

191191Keep per-request text in the newest user turn, after the last cache breakpoint, and make every cache-invalidating change at a natural break, such as the first request after a compaction. The cache is a byte-exact prefix match over the request in order (tools, then system prompt, then messages), so a change anywhere invalidates everything after it.

192192

193When something must change mid-conversation, send it in the form that keeps the cache, where the model supports it. In Anthropic’s test on Claude Fable 5.1, Claude Opus 5.5, and Claude Sonnet 5.5, on a cached conversation of about 100,000 tokens, editing the system prompt cost about 25 to 50 times what the same text cost as a mid-conversation system message. Changing effort at the top level cost 20 to 40 times what it cost per message[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs):

193When something must change mid-conversation, send it in the form that keeps the cache, where the model supports it. In Anthropic’s test on Claude Fable 5.1, Claude Opus 5.5, and Claude Sonnet 5.5, on a cached conversation of about 100,000 tokens, editing the system prompt cost about 25 to 50 times what the same text cost as a mid-conversation system message. Changing effort at the top level cost 20 to 40 times what it cost per message[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs):

194194

195195<Frame>

196196 ![Bar chart: system prompt edit up to $1.26 USD, top-level effort change up to $1.02 USD, either in a message up to $0.026 USD](https://platform.claude.com/docs/images/cost-intel/cache-break-cost.svg)

197197</Frame>

198198

199At list prices, the stakes are highest on Claude Fable 5.1 and Claude Mythos 5.1, where a break rewrites the prefix at 1.25x the input price instead of reading it at 0.025x. On Claude Haiku 5.5, a system prompt edit cost 12.5x as much as an unchanged request, because the edit rewrote the whole conversation. On models that run the [prefix check](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#prefix-check), a request that makes this edit and sends earlier thinking blocks back can get a 400 error instead[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). The top-level effort change, in a conversation that held the model’s first reply with its thinking, rewrote all but the system prompt. These figures are from the Messages API.

199At list prices, the stakes are highest on Claude Fable 5.1 and Claude Mythos 5.1, where a break rewrites the prefix at 1.25x the input price instead of reading it at 0.025x. On Claude Haiku 5.5, a system prompt edit cost 12.5x as much as an unchanged request, because the edit rewrote the whole conversation. On models that run the [prefix check](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#prefix-check), a request that makes this edit and sends earlier thinking blocks back can get a 400 error instead[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). The top-level effort change, in a conversation that held the model’s first reply with its thinking, rewrote all but the system prompt. These figures are from the Messages API.

200200

201201**What to do.**

202202

203* Put nothing that changes per request, such as a timestamp or a queue position, ahead of the stable prefix: it turns every request into a full cache write. In Anthropic’s test, a 25-token status line at the front of the system prompt made each request write the whole conversation; moved to the end of the newest user turn, it left the cache read intact[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

203* Put nothing that changes per request, such as a timestamp or a queue position, ahead of the stable prefix: it turns every request into a full cache write. In Anthropic’s test, a 25-token status line at the front of the system prompt made each request write the whole conversation; moved to the end of the newest user turn, it left the cache read intact[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

204204* The [prompt caching](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#what-invalidates-the-cache) page lists the changes that invalidate the cache, such as a change to [`effort`](https://platform.claude.com/docs/en/build-with-claude/effort), the thinking configuration, the [`speed`](https://platform.claude.com/docs/en/build-with-claude/fast-mode) setting, the system prompt, or a tool definition. Changing the output format invalidates the cache for that conversation (see [structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs#prompt-modification-and-token-costs)).

205205* Check which models support each form: [mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages), and [per-message effort](https://platform.claude.com/docs/en/build-with-claude/effort#changing-effort-mid-conversation), which is in beta.

206206* After a cache-invalidating change, confirm cache reads have not dropped; if they have, [cache diagnostics](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics) shows where the prefix diverged.

207207

208On an earlier model, Anthropic measured when to make such a change, on the triage agent’s long sessions[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). An effort change and an added tool cost more mid-session, or on the request that triggered the compaction, than on the first request after compaction. Accuracy was within run-to-run noise. The sessions with no changes cost more than those changed after compaction for another reason: they ran at the default effort throughout, and two of them compacted a second time, while the others started at low effort:

208On an earlier model, Anthropic measured when to make such a change, on the triage agent’s long sessions[18](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). An effort change and an added tool cost more mid-session, or on the request that triggered the compaction, than on the first request after compaction. Accuracy was within run-to-run noise. The sessions with no changes cost more than those changed after compaction for another reason: they ran at the default effort throughout, and two of them compacted a second time, while the others started at low effort:

209209

210210<Frame>

211211 ![Bar chart, cost per triage session: $0.81 USD no changes, $0.95 USD mid-session, $0.92 USD at compaction, $0.75 USD after](https://platform.claude.com/docs/images/cost-intel/compaction-timing.svg)

from line 227

227227

228228The levers interact with the cache and each other, so judge them by net effect, and use [cache diagnostics](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics) to confirm your cached prefix survives each change.

229229

230In measurements on an earlier model[28](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Anthropic ran them on an issue-triage agent working through 20 real bug reports with screenshots from a public repository, and on a longer variant of the same job with 2.6 times the tokens. With caching on, input trimming (image resizing and tool search) cut the cost of both runs further.

230In measurements on an earlier model[28](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Anthropic ran them on an issue-triage agent working through 20 real bug reports with screenshots from a public repository, and on a longer variant of the same job with 2.6 times the tokens. With caching on, input trimming (image resizing and tool search) cut the cost of both runs further.

231231

232232#### Defer unused tool definitions

233233

234234Every tool definition attached to a request is input on every turn, and a few MCP servers add up to hundreds of them.

235235

236In a measurement on an earlier model[27](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Anthropic ran the triage agent with its own two tools plus a catalog of real tool definitions from public MCP servers, for a total of up to 502 tools, loading all of them or marking the extras `defer_loading` behind [tool search](https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-search-tool):

236In a measurement on an earlier model[27](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Anthropic ran the triage agent with its own two tools plus a catalog of real tool definitions from public MCP servers, for a total of up to 502 tools, loading all of them or marking the extras `defer_loading` behind [tool search](https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-search-tool):

237237

238238<Frame>

239239 ![Line chart: with all tools loaded, run cost rises from $0.55 USD to $1.02 USD at 502 tools; tool search keeps it at $0.56 USD](https://platform.claude.com/docs/images/cost-intel/tool-search.svg)

from line 247

247247

248248#### Keep data files out of the prompt

249249

250When the model has to compute over a table, upload it with the [Files API](https://platform.claude.com/docs/en/build-with-claude/files) and let the model query it with [code execution](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool) instead of pasting it in. Anthropic asked 25 aggregate questions[15](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) (sums, filtered counts, group-bys, and a date filter) over a 1,862-row public CSV, with the answers computed by pandas:

250When the model has to compute over a table, upload it with the [Files API](https://platform.claude.com/docs/en/build-with-claude/files) and let the model query it with [code execution](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool) instead of pasting it in. Anthropic asked 25 aggregate questions[15](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) (sums, filtered counts, group-bys, and a date filter) over a 1,862-row public CSV, with the answers computed by pandas:

251251

252252<Frame>

253253 ![Scatter chart: file uploaded with code execution, 25 of 25 correct at $0.40 USD; pasted into the prompt, 7 of 25 at $5.01 USD](https://platform.claude.com/docs/images/cost-intel/data-files.svg)

from line 263

263263 ![Bar chart by run length: context editing adds 73% on the short run; compaction saves 32% and pruning 39% on the long](https://platform.claude.com/docs/images/cost-intel/hygiene.svg)

264264</Frame>

265265

266Anthropic measured these levers on an earlier model[28](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs):

266Anthropic measured these levers on an earlier model[28](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs):

267267

268268* On the 20-issue run the levers saved little or nothing, and context editing made the run cost more.

269269* On the long run the prune and compaction both cut the cost, while context editing changed nothing.

from line 359

359359

360360The command then proposes its edits as a diff (one hunk shown) and lists what it deliberately left alone: the refund window, the tone requirement, and the quality bar. You review a patch, not a rewrite.

361361

362Anthropic measured the effect on a support-desk evaluation[14](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), across two migrations between earlier models:

362Anthropic measured the effect on a support-desk evaluation[14](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), across two migrations between earlier models:

363363

364364<Frame>

365365 ![Scatter chart, support-desk evaluation: the old prompt costs more on the new model; audited, it is cheaper and as accurate](https://platform.claude.com/docs/images/cost-intel/prompt-audit.svg)

from line 388

388388

389389Compare models on cost per completed task, not per token, because you pay for completed tasks. For most agent workloads, start with Claude Opus 5.5 at its default effort (`medium`).

390390

391A more capable model can finish a task with less work, which can offset much of the per-token premium. Anthropic measured this on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset, whose scores are not comparable to the public leaderboard, priced as a customer is billed:

391A more capable model can finish a task with less work, which can offset much of the per-token premium. Anthropic measured this on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset, whose scores are not comparable to the public leaderboard, priced as a customer is billed:

392392

393393<Frame>

394394 ![Scatter chart, SWE-bench Pro: Claude Opus 5.5 at low effort matches Sonnet 5.5's default for about 7% more per solved task](https://platform.claude.com/docs/images/cost-intel/cost-per-task-sonnet-5-5.svg)

from line 396

396396

397397The following table lists the results:

398398

400| ----------------------------------------------------------------------------------------------------------------------------------------------- | ----------------- | ------------------ | --------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- |

407| GPQA Diamond[9](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Haiku 5.5 | `medium` (default) | 85% accuracy, compared with 91% for each of Claude Opus 5.5 and Claude Sonnet 5.5 | About a twentieth of Claude Opus 5.5’s cost per correct answer and about a tenth of Claude Sonnet 5.5’s |

400| ---------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------- | ------------------ | --------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- |

407| GPQA Diamond[9](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Haiku 5.5 | `medium` (default) | 85% accuracy, compared with 91% for each of Claude Opus 5.5 and Claude Sonnet 5.5 | About a twentieth of Claude Opus 5.5’s cost per correct answer and about a tenth of Claude Sonnet 5.5’s |

409409

410410**What to do.**

411411

from line 412

412412* Start with Claude Opus 5.5 at its default effort: on the SWE-bench Pro subset it matched Claude Fable 5.1 at its default, inside run-to-run noise, for about a fifth of the cost per solved task. Opus 5.5 and Fable 5.1 both largely saturate the SWE-bench Pro subset. Use Claude Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your evals on Claude Opus 5.5 at higher effort still fall short.

413413* At the other end, Claude Haiku 5.5 suits high-volume and latency-sensitive work with checkable outputs.

414414* Price every candidate in cost per completed task on your own traffic, including Claude Opus 5.5 at its default effort and the frontier model at reduced effort. Each of Claude Opus 5.5’s token prices, cache reads included, is twice Claude Sonnet 5.5’s, but on the SWE-bench Pro subset Opus 5.5 at `low` solved about the same share of tasks as Sonnet 5.5 at its default, inside run-to-run noise, for about 7% more per solved task.

415* Measure what a more expensive model buys on your own workload. On the 45 DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) tasks scored for every model, at default effort, Claude Opus 5.5 scored 2.6 points above Claude Sonnet 5.5 for 1.5 times the cost per task, and Claude Fable 5.1 scored 6.0 points above Claude Sonnet 5.5 for 5.8 times. On all 50 tasks, with four biology tasks that only Claude Sonnet 5.5 answered counted as failures, neither is measurably ahead of Claude Sonnet 5.5.

415* Measure what a more expensive model buys on your own workload. On the 45 DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) tasks scored for every model, at default effort, Claude Opus 5.5 scored 2.6 points above Claude Sonnet 5.5 for 1.5 times the cost per task, and Claude Fable 5.1 scored 6.0 points above Claude Sonnet 5.5 for 5.8 times. On all 50 tasks, with four biology tasks that only Claude Sonnet 5.5 answered counted as failures, neither is measurably ahead of Claude Sonnet 5.5.

416416* Price the tail of your workload, not the median: compare models on the hardest tenth of your tasks, not the typical one.

417417

418The bill is decided by the tasks the cheaper model fails, because a failed task still bills its tokens, then the retry, then whatever the failure costs downstream. The tail is also where the money goes even when nothing fails. On a 20-problem WideSearch[1](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) run, two problems carried 43% of the spend:

418The bill is decided by the tasks the cheaper model fails, because a failed task still bills its tokens, then the retry, then whatever the failure costs downstream. The tail is also where the money goes even when nothing fails. On a 20-problem WideSearch[1](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) run, two problems carried 43% of the spend:

419419

420420<Frame>

421421 ![Bar chart of 20 WideSearch problems ranked by cost: the top two carry 43% of spend and the cheapest half 10%](https://platform.claude.com/docs/images/cost-intel/tail.svg)

from line 425

425425

426426### Upgrade the model

427427

428If you are a model or two behind, the cheapest lever is the model string. In Anthropic’s measurements, each newer model solved at least as many tasks as the one before it, usually for less per solved task. The chart plots recent Claude models on two benchmarks in which an agent works alone on one written task and tests check the result: the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset, where every model ran through the same test setup, and Terminal-Bench 3[20](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). The table lists what each upgrade changed:

428If you are a model or two behind, the cheapest lever is the model string. In Anthropic’s measurements, each newer model solved at least as many tasks as the one before it, usually for less per solved task. The chart plots recent Claude models on two benchmarks in which an agent works alone on one written task and tests check the result: the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset, where every model ran through the same test setup, and Terminal-Bench 3[20](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). The table lists what each upgrade changed:

429429

430430<Frame>

431431 ![Charts: on SWE-bench Pro, Opus 5.5 and Sonnet 5.5 cut cost per solved task most; on Terminal-Bench 3, Opus drops to $28 USD](https://platform.claude.com/docs/images/cost-intel/upgrade-ladder-sonnet-5-5.svg)

432432</Frame>

433433

435| ------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------ | --------------------------------------- | ----------------------------------------------------- | ------------------------------------------------------------------------------------------------ |

442| Claude Managed Agents task set[25](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Haiku 4.5 to Claude Haiku 5.5 | As served; `medium` on Claude Haiku 5.5 | From 62% to 93% of attempts passed, one run per model | From $0.041 USD to about $0.010 USD in token cost per attempt; about 84% less per passed attempt |

435| ------------------------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------ | --------------------------------------- | ----------------------------------------------------- | ------------------------------------------------------------------------------------------------ |

442| Claude Managed Agents task set[25](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Haiku 4.5 to Claude Haiku 5.5 | As served; `medium` on Claude Haiku 5.5 | From 62% to 93% of attempts passed, one run per model | From $0.041 USD to about $0.010 USD in token cost per attempt; about 84% less per passed attempt |

443443

444444**When it pays.**

445445

446446* The size of the saving depends on the work. From Claude Opus 5 to Claude Opus 5.5, it is 78% per solved task on the SWE-bench Pro subset and 40% on typical workloads ([Introducing Claude Opus 5.5](https://www.anthropic.com/news/claude-opus-5-5)). From Claude Sonnet 5 to Claude Sonnet 5.5, it is 87% on the same subset and up to 30% for most work ([Introducing Claude Sonnet 5.5](https://www.anthropic.com/news/claude-sonnet-5-5)).

447447* Where older models already solve most tasks, the cost of each attempt decides the saving: on the SWE-bench Pro subset, Claude Opus 5.5 scores about the same as Claude Opus 5 for 78% less per solved task.

448* Where older models mostly fail, the pass rate decides the saving: on Terminal-Bench 3[20](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), the upgrade from Claude Opus 4.8 to Claude Opus 5 raises tasks solved from 15% to 41% and costs 56% less per solved task.

448* Where older models mostly fail, the pass rate decides the saving: on Terminal-Bench 3[20](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), the upgrade from Claude Opus 4.8 to Claude Opus 5 raises tasks solved from 15% to 41% and costs 56% less per solved task.

449449* Compare on cost per solved task, not per token: the same text costs about 30% more tokens on Claude 4.7 and later models, so a per-token comparison makes the newer models look more expensive by construction.

450450

451451### Tune effort

from line 454

454454

455455The following table lists Anthropic’s measurements on two benchmarks with current models; scores and costs are not comparable between the two benchmarks. Costs are per task, not per solved task as in [Compare models on cost per task](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#compare-models-on-cost-per-task). Each score pools two runs on SWE-bench Pro and three on DeepResearch Bench II, unless its cell says otherwise:

456456

458| --------------------------------------------------------------------------------------------------------------------------- | ---------------- | ------------------ | ------------------------------------ | ------------- |

459| SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset | Claude Opus 5.5 | `low` | 87.4% of tasks solved | $0.11 USD |

463| DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Fable 5.1 | `low` | Rubric score 60.5% | $3.47 USD |

458| -------------------------------------------------------------------------------------------------------------------------------------- | ---------------- | ------------------ | ------------------------------------ | ------------- |

459| SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset | Claude Opus 5.5 | `low` | 87.4% of tasks solved | $0.11 USD |

463| DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Claude Fable 5.1 | `low` | Rubric score 60.5% | $3.47 USD |

466466

467The following chart plots accuracy against cost per task for Claude Opus 5.5 on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset:

467The following chart plots accuracy against cost per task for Claude Opus 5.5 on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset:

468468

469469<Frame>

470470 ![Line chart of accuracy against cost per task by effort: steep from low to high for Claude Opus 5.5 on SWE-bench Pro](https://platform.claude.com/docs/images/cost-intel/effort-sweep-opus-5-5-swe-bench-pro.svg)

471471</Frame>

472472

473The following chart plots rubric score against cost per task for Claude Fable 5.1 on DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs):

473The following chart plots rubric score against cost per task for Claude Fable 5.1 on DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs):

474474

475475<Frame>

476476 ![Line chart, DeepResearch Bench II, Fable 5.1: rubric score 60.5% at low effort, 60.5% at medium, 63.0% at high, as cost rises](https://platform.claude.com/docs/images/cost-intel/effort-limit-repriced.svg)

from line 479

479479**What to do.**

480480

481481* Sweep two or three effort levels on a sample of your own traffic, each in a separate session: changing top-level effort mid-session [invalidates the cache](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#cache-repeated-context) and distorts the comparison. The curve is the baseline any multi-model strategy must beat ([step 2 of measuring on your own workload](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#measure-on-your-own-workload)).

482* On long-horizon coding, effort is a real tradeoff, which [re-running failures at higher effort](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#re-run-failures-at-higher-effort) turns back into a saving. At `medium`, Claude Sonnet 5.5 scored about 7 points lower than at its default, `high`, on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset for about 59% of the cost per task (2 runs each), and scored 62.7% compared with 77.8% on the internal agentic-coding benchmark[11](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) (1 attempt per task).

482* On long-horizon coding, effort is a real tradeoff, which [re-running failures at higher effort](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#re-run-failures-at-higher-effort) turns back into a saving. At `medium`, Claude Sonnet 5.5 scored about 7 points lower than at its default, `high`, on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset for about 59% of the cost per task (2 runs each), and scored 62.7% compared with 77.8% on the internal agentic-coding benchmark[11](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) (1 attempt per task).

483483* At `xhigh` and `max`, Claude Sonnet 5.5 can start its own rounds of review after it finishes a task, which takes more time and tokens. To keep that work on the task, add the prompt under “Thoroughness at `xhigh` and `max` effort” in [Prompting Claude Sonnet 5.5](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-sonnet-5-5#steer-initiative-and-scope) to your system prompt.

484* On research tasks, measure the model you ship. Claude Fable 5.1 scored 2.5 points higher at `high` than at `low` on DeepResearch Bench II, for 1.6 times the cost, and no higher at `medium` than at `low`; Claude Opus 5.5 scored 13.3 points lower at `low` than at `medium` on DRACO[21](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). The two results come from different models on different benchmarks, so they show that the answer varies, not why.

485* At `low`, Claude Opus 5.5’s typical task took 84% less time than at `medium` on DRACO[21](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), 30% less on HLE[22](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), and 47% less on an internal set of physics problems[23](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), and it scored 13.3, 3.4, and 14.0 points lower (3, 3, and 4 attempts per task).

486* On an internal defect-search benchmark for work larger than one context window[8](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Claude Fable 5.1 took 15.2, 17.5, and 19.9 hours per run at `low`, `medium`, and `high` (3 runs each). To save time, also compare lower effort with the two changes in [Show the model elapsed time](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#show-the-model-elapsed-time): `low` effort saved more time, and the changes kept more of the score.

484* On research tasks, measure the model you ship. Claude Fable 5.1 scored 2.5 points higher at `high` than at `low` on DeepResearch Bench II, for 1.6 times the cost, and no higher at `medium` than at `low`; Claude Opus 5.5 scored 13.3 points lower at `low` than at `medium` on DRACO[21](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). The two results come from different models on different benchmarks, so they show that the answer varies, not why.

485* At `low`, Claude Opus 5.5’s typical task took 84% less time than at `medium` on DRACO[21](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), 30% less on HLE[22](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), and 47% less on an internal set of physics problems[23](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), and it scored 13.3, 3.4, and 14.0 points lower (3, 3, and 4 attempts per task).

486* On an internal defect-search benchmark for work larger than one context window[8](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Claude Fable 5.1 took 15.2, 17.5, and 19.9 hours per run at `low`, `medium`, and `high` (3 runs each). To save time, also compare lower effort with the two changes in [Show the model elapsed time](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#show-the-model-elapsed-time): `low` effort saved more time, and the changes kept more of the score.

487487

488488### Re-run failures at higher effort

489489

490490When a task’s outcome is checkable, the cheapest policy on the effort curve is not a fixed setting: run every task at a low setting and re-run only the failures at a higher one.

491491

492Anthropic computed this policy task by task from the Claude Opus 5.5 effort runs on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset in [Tune effort](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#tune-effort):

492Anthropic computed this policy task by task from the Claude Opus 5.5 effort runs on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset in [Tune effort](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#tune-effort):

493493

494494<Frame>

495495 ![Chart, SWE-bench Pro, Opus 5.5: low or medium effort with failures re-run at high matches any fixed effort for less than high](https://platform.claude.com/docs/images/cost-intel/escalation-opus-5-5.svg)

from line 518

518518* Set `max_tokens` to 64,000 for agentic work, or to 128,000, the maximum, when a single cut-off attempt is costly. [Stream responses](https://platform.claude.com/docs/en/build-with-claude/streaming) that large, treat [`stop_reason: max_tokens`](https://platform.claude.com/docs/en/build-with-claude/handling-stop-reasons#max-tokens) as a failure, and save money with effort, which the model can see.

519519* On Claude Haiku 5.5, raise a small `max_tokens` or lower effort: thinking is on by default and counts toward the cap, so a response can end before any text (see the [Claude Haiku 5.5 migration guide](https://platform.claude.com/docs/en/models/haiku-5-5/migration-guide#configure-thinking)).

520520

521The following table shows what each model solved and what a solved task cost at each cap on an internal repository-task benchmark[12](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), one attempt per task, each model at its default effort. The rows at 16,384 average two runs, and the other rows are one run each:

521The following table shows what each model solved and what a solved task cost at each cap on an internal repository-task benchmark[12](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), one attempt per task, each model at its default effort. The rows at 16,384 average two runs, and the other rows are one run each:

522522

524524| ---------------- | ------------ | ------------------------- | ------------ | ------------------------ |

from line 533

533533

534534Compare the caps within each model. The two models were run a month apart, and their costs are priced on different bases.

535535

536On the internal repository tasks[12](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), almost every turn finishes far below a 16,384-token cap. The rare long turn is what a higher cap buys:

536On the internal repository tasks[12](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), almost every turn finishes far below a 16,384-token cap. The rare long turn is what a higher cap buys:

537537

538538<Frame>

539539 ![Dot plot of per-turn output for Opus 5.5 and Fable 5.1: medians a few hundred tokens, longest 61k and 128k, against the caps](https://platform.claude.com/docs/images/cost-intel/max-tokens-ladder-opus-5-5.svg)

from line 541

541541

542542Ask for shorter answers. Output tokens cost five times input tokens on Claude Haiku 5.5, Claude Sonnet 5.5, Claude Opus 5.5, and Claude Fable 5.1, and in an agent loop every token the model writes comes back as input on every later turn, so you pay for a long answer again and again.

543543

544In a measurement on an earlier model[26](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Anthropic ran a small agent that triages bug reports ([Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens)) under three final-answer instructions, three runs each, with the same model and tools:

544In a measurement on an earlier model[26](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Anthropic ran a small agent that triages bug reports ([Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens)) under three final-answer instructions, three runs each, with the same model and tools:

545545

546546* The original asked for two lines.

547547* The shorter variant asked for one.

from line 557

557557

558558### Show the model elapsed time

559559

560To make an agent finish sooner and cost less, tell it that time matters, and show it the elapsed time on every request. In Anthropic’s measurement, the two changes cut the time of a single Claude Opus 5.5 agent at `medium` effort by 47% and its cost by 60% on the typical (median) DRACO[21](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) research task, and its average score was 4.1 points lower.

560To make an agent finish sooner and cost less, tell it that time matters, and show it the elapsed time on every request. In Anthropic’s measurement, the two changes cut the time of a single Claude Opus 5.5 agent at `medium` effort by 47% and its cost by 60% on the typical (median) DRACO[21](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) research task, and its average score was 4.1 points lower.

561561

562On HLE[22](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) and an internal physics set[23](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), the typical task took 14% and 7% less time, and the physics score was 3.0 points lower. Lowering effort to `low` saved more time and cost than the changes did, and lost more score (reference 21). Check the score on your own tasks before you adopt the changes.

562On HLE[22](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) and an internal physics set[23](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), the typical task took 14% and 7% less time, and the physics score was 3.0 points lower. Lowering effort to `low` saved more time and cost than the changes did, and lost more score (reference 21). Check the score on your own tasks before you adopt the changes.

563563

564564For working code, see the Claude Cookbook recipe [Multi-agent teams under latency pressure and budgets](https://github.com/anthropics/claude-cookbooks/blob/main/patterns/agents/latency_multi_agent.ipynb), which applies both changes to a team of agents. To say that time matters in the system prompt instead, see [Time signals for multiagent harnesses](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5-5#time-signals-for-multi-agent-harnesses).

565565

from line 596

596596

597597The following table compares each pairing with its executor alone at the same effort:

598598

600| ---------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------- | -------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------- |

601| Internal agentic-coding benchmark[11](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | 90.1%, 1.7 points over the executor alone, at the edge of run-to-run noise | $2.92 USD per attempt, about 2.1 times the executor alone | About 1.4 times per attempt; 4% of attempts received no advice |

602| GPQA Diamond[9](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Within run-to-run noise of the executor alone | 12% more per question for Claude Haiku 5.5, 25% more for Claude Sonnet 5.5 | On none of the 198 questions, in two runs each; no system prompt asked for consultations |

603| Chartography[13](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | 61.7, 7 points below the executor alone and beyond run-to-run noise | About the same as the executor alone | On 1 of 300 tasks |

600| --------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------- | -------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------- |

601| Internal agentic-coding benchmark[11](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | 90.1%, 1.7 points over the executor alone, at the edge of run-to-run noise | $2.92 USD per attempt, about 2.1 times the executor alone | About 1.4 times per attempt; 4% of attempts received no advice |

602| GPQA Diamond[9](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Within run-to-run noise of the executor alone | 12% more per question for Claude Haiku 5.5, 25% more for Claude Sonnet 5.5 | On none of the 198 questions, in two runs each; no system prompt asked for consultations |

603| Chartography[13](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | 61.7, 7 points below the executor alone and beyond run-to-run noise | About the same as the executor alone | On 1 of 300 tasks |

604604

605605* **Internal agentic-coding benchmark.** Claude Opus 5.5 at `high`, with a Claude Fable 5.1 advisor.

606606* **GPQA Diamond.** Claude Haiku 5.5 or Claude Sonnet 5.5 at each model’s default (`medium` and `high`), with a Claude Opus 5.5 advisor.

from line 617

617617**What to do.**

618618

619619* Measure the consult rate on your own tasks. The advisor tool’s definition, about 1,000 prompt tokens per request, adds cost even when the executor never consults. With [prompt caching](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#cache-repeated-context), a multi-turn agent loop pays for the definition at the cache-read price after the first request, so this overhead matters most for short, single-request tasks.

620* If the executor rarely consults, see [Prompting for coding and agent tasks](https://platform.claude.com/docs/en/agents-and-tools/tool-use/advisor-tool#prompting-for-coding-and-agent-tasks) for system prompts that ask for consultations, and measure again after you add one. On 20 SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) problems, the suggested prompt made no clear difference for Claude Sonnet 5.5.

621* Restore the executor’s effort if the consult rate collapses. In measurements on earlier models, one pairing that consulted on most tasks at the default effort consulted on few when effort was lowered, and then scored below the executor alone at its default effort (SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs)).

620* If the executor rarely consults, see [Prompting for coding and agent tasks](https://platform.claude.com/docs/en/agents-and-tools/tool-use/advisor-tool#prompting-for-coding-and-agent-tasks) for system prompts that ask for consultations, and measure again after you add one. On 20 SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) problems, the suggested prompt made no clear difference for Claude Sonnet 5.5.

621* Restore the executor’s effort if the consult rate collapses. In measurements on earlier models, one pairing that consulted on most tasks at the default effort consulted on few when effort was lowered, and then scored below the executor alone at its default effort (SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs)).

622622* Recheck at every model release, because releases move both the capability gap and the price ratio.

623623

624624### Orchestrator strategy: delegate bulk work

from line 625

625625

626626Delegation saved money in only two situations, in measurements with workers on Claude Sonnet 5, an earlier model:

627627

628* Routine work with a long cost tail (an easy slice of BrowseComp[4](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs))

629* Work larger than one context window (an internal defect-search benchmark[8](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs))

628* Routine work with a long cost tail (an easy slice of BrowseComp[4](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs))

629* Work larger than one context window (an internal defect-search benchmark[8](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs))

630630

631When the work is one dependent chain, or fits in a single context without a long cost tail, the orchestrator’s model alone at lower effort came out ahead in every case measured on earlier models (the harder BrowseComp set and DeepWideSearch[6](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs)); independent external work reports the same pattern[5](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

631When the work is one dependent chain, or fits in a single context without a long cost tail, the orchestrator’s model alone at lower effort came out ahead in every case measured on earlier models (the harder BrowseComp set and DeepWideSearch[6](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs)); independent external work reports the same pattern[5](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

632632

633633To build an orchestrator, use [multiagent orchestration](https://platform.claude.com/docs/en/managed-agents/multiagent-orchestration) in Claude Managed Agents: configure an orchestrator agent and worker agents that it can call, each with its own model.

634634

from line 645

645645**What to do.**

646646

647647* **Routine work with a long cost tail:** measure this case first, for accuracy as well as cost. The measurement behind it covers only problems that the delegated configuration answered correctly in every run (reference 4).

648* **Work larger than one context window:** delegate partitions to workers. Lowering effort cannot help, because the bill is the reading itself: on the defect-search benchmark[8](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Claude Fable 5.1 alone cost $468 USD to $552 USD per run across its three effort settings and took 15 to 20 hours. With earlier-model workers, delegation cost less and finished sooner, for a lower score.

649* **Claude Haiku 5.5 subagents on coding tasks:** compare them with lower effort first. On an internal coding benchmark on Claude Managed Agents, they were an alternative to lowering effort, not better. Adding them to Claude Opus 5.5 at `medium`, with a short delegation prompt, cut cost per attempt by 32% and the pass rate by about 4 percentage points (95% interval 1.3 to 7.1), about what a mix of `low` and `medium` without subagents would give at that cost[29](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

648* **Work larger than one context window:** delegate partitions to workers. Lowering effort cannot help, because the bill is the reading itself: on the defect-search benchmark[8](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Claude Fable 5.1 alone cost $468 USD to $552 USD per run across its three effort settings and took 15 to 20 hours. With earlier-model workers, delegation cost less and finished sooner, for a lower score.

649* **Claude Haiku 5.5 subagents on coding tasks:** compare them with lower effort first. On an internal coding benchmark on Claude Managed Agents, they were an alternative to lowering effort, not better. Adding them to Claude Opus 5.5 at `medium`, with a short delegation prompt, cut cost per attempt by 32% and the pass rate by about 4 percentage points (95% interval 1.3 to 7.1), about what a mix of `low` and `medium` without subagents would give at that cost[29](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs).

650650

651651## Measure on your own workload

652652

from line 853

853853

854854The following table lists the levers in the order to try them. The figures come from Anthropic’s tests, launch posts, and published prices and speeds:

855855

857| -------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

858| Prompt caching | Cost cut by a factor of 4.8 to 7.2 on a research agent, and by 83% to 86% on a small triage agent | None | Faster | [Cache repeated context](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#cache-repeated-context) |

859| 1-hour cache duration | With 1 turn in 10 after a pause, the 5-minute default cost 54% more; with 1 in 20, about the same | None | On earlier models, stayed warm after a pause[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | [Pick the cache duration](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#pick-the-cache-duration) |

860| Input trimming | A further 21% to 26% off a small triage agent’s bill | With tool search, accuracy in the same range as without[27](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Neutral | [Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens) |

861| Tool search | 45% with 500 tool definitions attached; 20% from deferring a GitHub MCP server’s toolset | None for tools the task never needs | Neutral | [Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens) |

862| Data files through code execution | 92% on a data task of 25 questions, with prompt caching off | A gain: 25 of 25 correct instead of 7 | Faster | [Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens) |

864| Prompt audit against the current model | 14% on both model upgrades tested | None; a gain on one | Faster (fewer tool rounds) | [Audit prompts against the current model](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#audit-prompts-against-the-current-model) |

865| Upgrade the model | Usually less: up to 40% less on typical workloads, and up to 87% less per solved task on a coding benchmark | None; a gain on most | Neutral | [Upgrade the model](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#upgrade-the-model) |

866| Lower effort | On long coding tasks, `medium` cost about 30% to 52% less than `high`, and `low` about two thirds less | Up to 15 points lower, depending on the task; 2.5 on one research agent | Faster | [Tune effort](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#tune-effort) |

867| Re-run failures | About 40% less than running everything at `high`, at the same pass rate or slightly better | None | A second attempt on the tasks that fail | [Re-run failures at higher effort](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#re-run-failures-at-higher-effort) |

869| Raising `max_tokens` | None: cost per solved task stayed about the same | A gain: up to 22 points more tasks solved on one coding benchmark; none on another | Neutral | [Set budgets and output caps](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#set-budgets-and-output-caps) |

870| Advisor | None: about 2.1 times the cost of the model alone, on a coding benchmark | On coding, a small gain that may be chance; on chart reading, a loss | About 1.4 extra calls per attempt, on coding | [Advisor strategy](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#advisor-strategy-escalate-hard-decisions) |

871| Orchestrator | About half the cost of the orchestrator’s model alone, on work larger than one context window; on routine work, measure first | 10 to 12 points lower on that work | Much faster on that work | [Orchestrator strategy](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#orchestrator-strategy-delegate-bulk-work) |

872| Fast mode | None: twice the standard token prices | None | Up to 2.5x higher output tokens per second | [Show the model elapsed time](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#show-the-model-elapsed-time) |

857| -------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

859| 1-hour cache duration | With 1 turn in 10 after a pause, the 5-minute default cost 54% more; with 1 in 20, about the same | None | On earlier models, stayed warm after a pause[16](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | [Pick the cache duration](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#pick-the-cache-duration) |

860| Input trimming | A further 21% to 26% off a small triage agent’s bill | With tool search, accuracy in the same range as without[27](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) | Neutral | [Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens) |

873873

874874### Automate the search with the Claude API skill

875875

from line 889

889889/claude-api hillclimb

890890```

891891

892Anthropic measured the first lever of `cost-optimize` and two searches on tau2-bench retail[24](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), all on earlier models. The cache breakpoints that `cost-optimize` produced cut cost per task, and the pass rate fell by less than the spread of identical runs.

892Anthropic measured the first lever of `cost-optimize` and two searches on tau2-bench retail[24](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), all on earlier models. The cache breakpoints that `cost-optimize` produced cut cost per task, and the pass rate fell by less than the spread of identical runs.

893893

894894A search found a cheaper configuration by lowering effort and moving down a model tier. Its pass rate stayed inside the band set before the search by a fraction of a task, and ended below the stricter floor the search used for its own verdicts. A second search, started from the lower-cost model at its default effort, kept nothing. Reference 24 has the figures and what each search cost to run.

895895

agents-and-tools/mcp-tunnels/console Changed · +23 / -15 lines

from line 116

116116

117117## Use the tunnel in an agent

118118

119Once your stack is running and has one or more MCP servers configured, attach an upstream MCP server to a Managed Agent session. To call the same servers from the Messages API instead, see [Use the tunneled MCP servers](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/overview#use-the-tunneled-mcp-servers).

119Once your stack is running, add a tunneled MCP server to an agent when you create a [Claude Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) session. To call the same servers from the Messages API instead, see [Use the tunneled MCP servers](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/overview#use-the-tunneled-mcp-servers).

120120

121<Note>

122 The picker only shows tunnels with at least one active certificate. A tunnel that still shows **Needs certificate** in the **MCP tunnels** list does not appear in the dropdown menu; register a CA certificate first. The picker is also workspace-scoped: it lists tunnels in the same workspace as the session, not other workspaces.

123</Note>

124

125121<Steps>

126122 <Step title="Open the Create session dialog box">

127 In the Claude Console, go to the [Sessions](https://platform.claude.com/sessions) page (`platform.claude.com/sessions`) and click **Create session**.

123 In the Claude Console, go to the [Sessions](https://platform.claude.com/sessions) page (`platform.claude.com/sessions`). In the workspace picker in the sidebar, select the workspace that holds the tunnel. Click **Create session**.

128124 </Step>

129125

130 <Step title="Define an inline agent">

131 In the agent picker, select **Create new agent** so you can edit the MCP server list directly.

126 <Step title="Create a new agent">

127 From the **Agent** dropdown menu, select **Create new agent**, because the dialog box can add MCP servers only to a new agent. Enter a name for the agent in the **Name** field.

132128 </Step>

133129

134130 <Step title="Add the MCP server">

135 Click **+ MCP Server** and open the dropdown menu. Tunnels created in the current workspace appear at the top of the list, above the public connector catalog. Select the tunnel that fronts the server you want to reach.

131 Under **MCP servers**, click **MCP server**. On the card that appears, open the **Select a server** dropdown menu and select your tunnel from the top of the list. Only tunnels in the selected workspace that have an [active CA certificate](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/console#add-a-ca-certificate) are listed.

136132 </Step>

137133

138 <Step title="Supply the routing">

139 The card shows two optional fields: **Subdomain** (prefixed to the tunnel domain) and **Path** (appended after it). Fill in one or both, depending on how your proxy’s routes are configured. The **Resolves to** line shows the full MCP server URL that the agent connects to.

134 <Step title="Enter the subdomain and path">

135 In the **Subdomain** field, enter the subdomain that your proxy’s `routes` map sends to the MCP server, such as `wiki`. In the **Path** field, enter the path that the MCP server expects, such as `mcp`. Check that the **Resolves to** line shows your subdomain and path under the tunnel domain, such as `https://wiki.abcd1234.tunnel.anthropic.com/mcp`.

140136 </Step>

141</Steps>

142137

143<Note>

144 The tunnel carries traffic; it does not authenticate to the upstream MCP server. Configure OAuth or bearer authentication on the MCP server the same way as for any other MCP server.

145</Note>

138 <Step title="Add a credential if the MCP server needs one">

139 The tunnel carries traffic to the upstream MCP server but does not authenticate to it. If the server requires OAuth or a bearer token, click **Connect** on the card and save a credential in a vault. If the card shows **Needs vault** instead, go to **Credential vaults** and add a vault that holds a credential for the **Resolves to** URL. If the server requires no authentication, ignore **Connect** or **Needs vault** on the card. To learn how vaults work, see [Authenticate with vaults](https://platform.claude.com/docs/en/managed-agents/vaults).

140 </Step>

141

142 <Step title="Select an environment">

143 In the **Environment** field, above **MCP servers**, select an environment. If no warning appears below the field, go to the next step. If a warning says that the environment blocks MCP servers, click **Edit environment**, which opens the environment in a new browser tab. Under **Networking**, add the host from the **Resolves to** line, such as `wiki.abcd1234.tunnel.anthropic.com`, to **Allowed hosts**. Leave the **Allow MCP servers** checkbox unchecked, as the [Limit the network](https://platform.claude.com/docs/en/managed-agents/secure-deployment#limit-the-network) section recommends. Click **Save** and return to the browser tab with the dialog box.

144 </Step>

145

146 <Step title="Create the session">

147 Under **Credential vaults**, remove any vault that the session does not need. If a vault remains, select the checkbox that confirms you are authorized to use it. In the dialog box, click **Create session**. The Console saves the agent and opens the session.

148 </Step>

149

150 <Step title="Test the connection">

151 Ask the agent to use one of the MCP server’s tools. You should see a tool call followed by its result. If the agent can’t reach the server, see [Troubleshoot MCP tunnels](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/troubleshooting).

152 </Step>

153</Steps>

146154

147155## Archive a tunnel

148156

agents-and-tools/mcp-tunnels/overview Changed · +4 / -2 lines

from line 47

4747* **[cloudflared](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/concepts#components):** Cloudflare’s open-source tunnel connector. It initiates outbound-only connections to the [tunnel edge](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/concepts#components) and carries encrypted traffic from Anthropic to your proxy.

4848* **[Proxy](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/concepts#components):** Anthropic’s routing component. It terminates [inner TLS](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/concepts#components), validates that upstream IPs fall within an allowed range, and routes each request to the correct [upstream MCP server](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/concepts#components) based on hostname.

4949

50Each MCP server you expose gets a hostname under your tunnel domain (for example, `docs.<your-tunnel-domain>`). You attach these hostnames to a Managed Agent session in the Claude Console, or pass them to the Messages API through the [MCP connector](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector).

50Each MCP server you expose gets a hostname under your tunnel domain (for example, `docs.<your-tunnel-domain>`). You attach these hostnames to an agent in the Claude Console, or pass them to the Messages API through the [MCP connector](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector).

5151

5252## Prerequisites

5353

from line 158

158158### Managed Agents (Console)

159159

1601601. On the [Sessions](https://platform.claude.com/sessions) page (`platform.claude.com/sessions`) in the Claude Console, create a session and select **Create new agent** so you can edit the MCP server list.

1612. Click **+ MCP Server** and open the dropdown menu. Tunnels in the session’s workspace that have at least one active certificate appear at the top of the list, above the public connector catalog.

1612. Under **MCP servers**, click **MCP server** and open the dropdown menu. Tunnels in the session’s workspace that have at least one active certificate appear at the top of the list, above the public connector catalog.

1621623. Select the tunnel and supply the **Subdomain** that your proxy routes to a specific MCP server, and the **Path** the upstream MCP server expects. The **Resolves to** line shows the exact URL.

163

164For the full procedure, see [Use the tunnel in an agent](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/console#use-the-tunnel-in-an-agent).

163165

164166### Messages API

165167

agents-and-tools/tool-use/web-fetch-tool Changed · +4 / -5 lines

from line 495

495495

496496### Content limits

497497

498The `max_content_tokens` parameter limits the amount of content included in the context. If the fetched content exceeds this limit, the tool truncates it. This helps control token usage when fetching large documents. The limit applies to text content, not to binary content such as PDFs.

498The `max_content_tokens` parameter limits the amount of content included in the context. If the fetched content exceeds this limit, the tool truncates it. This helps control token usage when fetching large documents. The limit is approximate: the actual number of input tokens used can vary by a small amount. The limit applies to text content, not to binary content such as PDFs.

499499

500<Note>

501 The `max_content_tokens` parameter limit is approximate. The actual number of input tokens used can vary by a small amount.

502</Note>

500If a fetched PDF would make the request exceed the [PDF page or size limits](https://platform.claude.com/docs/en/build-with-claude/pdf-support#check-pdf-requirements), the `web_fetch_tool_result` is an error with the `content_too_large` error code. PDFs and images already in the conversation count toward those limits, so removing them can let the fetch succeed.

503501

504502On Claude Managed Agents, the `web_fetch` entry of the agent toolset also accepts `max_content_tokens`; see the [web tool settings](https://platform.claude.com/docs/en/managed-agents/tools-web-restrictions#settings).

505503

from line 682

684682* `url_not_accessible`: Failed to fetch content (HTTP error)

685683* `too_many_requests`: Rate limit exceeded

686684* `unsupported_content_type`: Content type not supported (only text, HTML, and PDF)

685* `content_too_large`: Fetched PDF would make the request exceed the PDF page or size limits (see [Content limits](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool#content-limits))

687686* `max_uses_exceeded`: Maximum web fetch tool uses exceeded

688687* `unavailable`: An internal error occurred

689688

api/beta Changed · +4994 / -222 lines

## Beta › Organization › Claude Tag › Slack Channels ### Set Up Slack Channel ### List Slack Channels ### Get Slack Channel ### Update Slack Channel ### Remove Slack Channel ## Beta › Organization › Claude Tag › Slack Workspaces ### List Slack Workspaces ### Get Slack Workspace ### Update Slack Workspace ## Beta › Organization › Claude Tag › Bundles ### List Bundles ### Get Bundle ## Beta › Organization › Claude Tag › Bundles › Targets ### List Bundle Targets ### Attach Bundle ### Get Bundle Target ### Detach Bundle

This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.

from line 9

1010### Anthropic Beta

1111

12- `AnthropicBeta = string or "message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 51 more`

12- `AnthropicBeta = string or "message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 52 more`

1313

14 Beta names you can send in the `anthropic-beta` header.

16 See [Beta headers](https://platform.claude.com/docs/en/api/beta-headers?sdk_languages=http) for how to send them and where to find the name a feature needs.

1418 - `string`

1519

1620 - `"message-batches-2024-09-24"`

from line 125

121125

122126 - `"analytics-usage-cost-2026-08-24"`

123127

128 - `"experimental-claude-tag-2026-10-21"`

129

124130### Beta API Error

125131

126132- `BetaAPIError object`

from line 611

605611

606612 - `"analytics-usage-cost-2026-08-24"`

607613

614 - `"experimental-claude-tag-2026-10-21"`

615

608616- `"anthropic-workspace-id": optional string`

609617

610618 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1100

10921100

10931101 - `"analytics-usage-cost-2026-08-24"`

10941102

1103 - `"experimental-claude-tag-2026-10-21"`

1104

10951105- `"anthropic-workspace-id": optional string`

10961106

10971107 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1568

15581568

15591569 - `"analytics-usage-cost-2026-08-24"`

15601570

1571 - `"experimental-claude-tag-2026-10-21"`

1572

15611573- `"anthropic-user-profile-id": optional string`

15621574

15631575 The user profile ID to attribute this request to. Use when acting on behalf of a party other than your organization.

from line 2469

24572469

24582470 - `BetaEncryptedCodeExecutionResultBlockParam object`

24592471

2460 Code execution result with encrypted stdout for PFC + web_search results.

2472 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form, sent back in a later request.

24612473

24622474 - `type: "encrypted_code_execution_result"`

24632475

from line 3005

29933005

29943006 - `BetaCodeExecutionTool20260120 object`

29953007

2996 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

3008 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

29973009

3010 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

3011

3012 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

3013

29983014 - `type: "code_execution_20260120"`

29993015

30003016 - `name: "code_execution"`

from line 5657

56415657

56425658 - `BetaCodeExecutionTool20260120 object`

56435659

5644 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

5660 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

56455661

5662 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

5663

5664 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

5665

56465666 - `BetaCodeExecutionTool20260521 object`

56475667

56485668 Code execution tool with REPL state persistence.

from line 6357

63376357

63386358 - `BetaEncryptedCodeExecutionResultBlock object`

63396359

6340 Code execution result with encrypted stdout for PFC + web_search results.

6360 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form.

63416361

6362 The API returns this block in place of `code_execution_result` when the code called the web search tool.

6363

6364 See the [programmatic tool calling guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling?sdk_languages=http) to learn more about programmatic tool calling.

6365

63426366 - `type: "encrypted_code_execution_result"`

63436367

63446368 default: encrypted_code_execution_result

from line 6923

68996923

69006924 - `BetaCodeExecutionTool20260120 object`

69016925

6902 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

6926 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

69036927

6928 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

6929

6930 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

6931

69046932 - `type: "code_execution_20260120"`

69056933

69066934 - `name: "code_execution"`

from line 7613

75857613 - `BetaComputerToolset20260801 object`

75867614

75877615 The computer toolset: a single `tools[]` entry (carrying no

7588 `name`) that declares the computer tool family. The model is

7589 served the family's tool with any members disabled via `configs`

7590 removed from its schema. Every member is enabled by default, zoom

7591 included. The single-tool options `display_number` and

7592 `enable_zoom` are not fields of a toolset entry — it carries only

7593 `type`, `configs`, and `cache_control`; zoom is controlled

7594 via `configs.zoom.enabled`.

7595

7596 - `type: "computer_toolset_20260801"`

7597

7598 - `cache_control: optional BetaCacheControlEphemeral or null`

7599

7600 Create a cache control breakpoint at this content block.

7601

7602 - `configs: optional BetaComputerToolsetConfigs or null`

7603

7604 Sparse per-member overrides, keyed by member name. Absent, null, and {} are equivalent; a member's defaults apply wherever its key is absent.

7605

7606 - `type: optional BetaComputerTypeConfig or null`

7607

7608 `type`'s config overrides.

7609

7610 - `defer_loading: optional boolean or null`

7611

7612 Defer loading for this member. Must resolve to the same value on every enabled member of the toolset.

7613

7614 - `enabled: optional boolean or null`

7615

7616 Whether this member is offered to the model. Default is per member, per the toolset's documentation. A member whose enabled resolves false is withheld from the served schema.

7617

7618 - `cursor_position: optional BetaComputerCursorPositionConfig or null`

7619

7620 `cursor_position`'s config overrides.

7621

7622 - `defer_loading: optional boolean or null`

7623

7624 Defer loading for this member. Must resolve to the same value on every enabled member of the toolset.

7625

7626 - `enabled: optional boolean or null`

7627

7628 Whether this member is offered to the model. Default is per member, per the toolset's documentation. A member whose enabled resolves false is withheld from the served schema.

7629

7630 - `double_click: optional BetaComputerDoubleClickConfig or null`

7631

7632 `double_click`'s config overrides.

7633

7634

7616 `name`) that declares the computer tool f

api/beta/agents Changed · +12 / -0 lines

This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.

from line 133

133133

134134 - `"analytics-usage-cost-2026-08-24"`

135135

136 - `"experimental-claude-tag-2026-10-21"`

137

136138- `"anthropic-workspace-id": optional string`

137139

138140 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2157

21552157

21562158 - `"analytics-usage-cost-2026-08-24"`

21572159

2160 - `"experimental-claude-tag-2026-10-21"`

2161

21582162- `"anthropic-workspace-id": optional string`

21592163

21602164 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 3215

32113215

32123216 - `"analytics-usage-cost-2026-08-24"`

32133217

3218 - `"experimental-claude-tag-2026-10-21"`

3219

32143220- `"anthropic-workspace-id": optional string`

32153221

32163222 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 4256

42504256

42514257 - `"analytics-usage-cost-2026-08-24"`

42524258

4259 - `"experimental-claude-tag-2026-10-21"`

4260

42534261- `"anthropic-workspace-id": optional string`

42544262

42554263 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 6255

62476255

62486256 - `"analytics-usage-cost-2026-08-24"`

62496257

6258 - `"experimental-claude-tag-2026-10-21"`

6259

62506260- `"anthropic-workspace-id": optional string`

62516261

62526262 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 8229

82198229

82208230 - `name: string`

82218231

8222 Name of the tool. Compared exactly, so upper and lower case letters are different.

8223

8224 minLength: 1, maxLength: 128

8225

8226 - `server_tool_results: BetaManagedAgentsWebFetchURLSourceToolFilter or null`

8227

8228 Which of the web_search and web_fetch t

8232 Name of the tool. Compared exactly, so u

api/beta/deployments Changed · +16 / -0 lines

from line 133

133133

134134 - `"analytics-usage-cost-2026-08-24"`

135135

136 - `"experimental-claude-tag-2026-10-21"`

137

136138- `"anthropic-workspace-id": optional string`

137139

138140 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1333

13311333

13321334 - `"analytics-usage-cost-2026-08-24"`

13331335

1336 - `"experimental-claude-tag-2026-10-21"`

1337

13341338- `"anthropic-workspace-id": optional string`

13351339

13361340 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2080

20762080

20772081 - `"analytics-usage-cost-2026-08-24"`

20782082

2083 - `"experimental-claude-tag-2026-10-21"`

2084

20792085- `"anthropic-workspace-id": optional string`

20802086

20812087 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2818

28122818

28132819 - `"analytics-usage-cost-2026-08-24"`

28142820

2821 - `"experimental-claude-tag-2026-10-21"`

2822

28152823- `"anthropic-workspace-id": optional string`

28162824

28172825 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 3966

39583966

39593967 - `"analytics-usage-cost-2026-08-24"`

39603968

3969 - `"experimental-claude-tag-2026-10-21"`

3970

39613971- `"anthropic-workspace-id": optional string`

39623972

39633973 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 4705

46954705

46964706 - `"analytics-usage-cost-2026-08-24"`

46974707

4708 - `"experimental-claude-tag-2026-10-21"`

4709

46984710- `"anthropic-workspace-id": optional string`

46994711

47004712 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 5106

50945106

50955107 - `"analytics-usage-cost-2026-08-24"`

50965108

5109 - `"experimental-claude-tag-2026-10-21"`

5110

50975111- `"anthropic-workspace-id": optional string`

50985112

50995113 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 5844

58305844 - `"message-threads-2026-08-12"`

58315845

58325846 - `"analytics-usage-cost-2026-08-24"`

5847

5848 - `"experimental-claude-tag-2026-10-21"`

58335849

58345850- `"anthropic-workspace-id": optional string`

58355851

api/beta/dreams Changed · +10 / -0 lines

from line 137

137137

138138 - `"analytics-usage-cost-2026-08-24"`

139139

140 - `"experimental-claude-tag-2026-10-21"`

141

140142- `"anthropic-workspace-id": optional string`

141143

142144 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 732

730732

731733 - `"analytics-usage-cost-2026-08-24"`

732734

735 - `"experimental-claude-tag-2026-10-21"`

736

733737- `"anthropic-workspace-id": optional string`

734738

735739 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1160

11561160

11571161 - `"analytics-usage-cost-2026-08-24"`

11581162

1163 - `"experimental-claude-tag-2026-10-21"`

1164

11591165- `"anthropic-workspace-id": optional string`

11601166

11611167 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1583

15771583

15781584 - `"analytics-usage-cost-2026-08-24"`

15791585

1586 - `"experimental-claude-tag-2026-10-21"`

1587

15801588- `"anthropic-workspace-id": optional string`

15811589

15821590 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2006

19982006 - `"message-threads-2026-08-12"`

19992007

20002008 - `"analytics-usage-cost-2026-08-24"`

2009

2010 - `"experimental-claude-tag-2026-10-21"`

20012011

20022012- `"anthropic-workspace-id": optional string`

20032013

api/beta/environments Changed · +30 / -4 lines

from line 133

133133

134134 - `"analytics-usage-cost-2026-08-24"`

135135

136 - `"experimental-claude-tag-2026-10-21"`

137

136138- `"anthropic-workspace-id": optional string`

137139

138140 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 618

616618

617619 - `"analytics-usage-cost-2026-08-24"`

618620

621 - `"experimental-claude-tag-2026-10-21"`

622

619623- `"anthropic-workspace-id": optional string`

620624

621625 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 959

955959

956960 - `"analytics-usage-cost-2026-08-24"`

957961

962 - `"experimental-claude-tag-2026-10-21"`

963

958964- `"anthropic-workspace-id": optional string`

959965

960966 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1291

12851291

12861292 - `"analytics-usage-cost-2026-08-24"`

12871293

1294 - `"experimental-claude-tag-2026-10-21"`

1295

12881296- `"anthropic-workspace-id": optional string`

12891297

12901298 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1745

17371745

17381746 - `"analytics-usage-cost-2026-08-24"`

17391747

1748 - `"experimental-claude-tag-2026-10-21"`

1749

17401750- `"anthropic-workspace-id": optional string`

17411751

17421752 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1920

19101920

19111921 - `"analytics-usage-cost-2026-08-24"`

19121922

1923 - `"experimental-claude-tag-2026-10-21"`

1924

19131925- `"anthropic-workspace-id": optional string`

19141926

19151927 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2722

27102722

27112723 - `"analytics-usage-cost-2026-08-24"`

27122724

2725 - `"experimental-claude-tag-2026-10-21"`

2726

27132727- `"anthropic-workspace-id": optional string`

27142728

27152729 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2994

29802994

29812995 - `"analytics-usage-cost-2026-08-24"`

29822996

2997 - `"experimental-claude-tag-2026-10-21"`

2998

29832999- `"Anthropic-Worker-ID": optional string`

29843000

29853001 Unique identifier for the specific worker polling, used to track aggregated environment-level work metrics in Console

from line 3252

32363252

32373253 - `"analytics-usage-cost-2026-08-24"`

32383254

3255 - `"experimental-claude-tag-2026-10-21"`

3256

32393257#### Returns

32403258

32413259- `BetaSelfHostedWork object`

from line 3517

34993517

35003518 - `"analytics-usage-cost-2026-08-24"`

35013519

3520 - `"experimental-claude-tag-2026-10-21"`

3521

35023522#### Returns

35033523

35043524- `BetaSelfHostedWorkHeartbeatResponse object`

from line 3533

35133533

35143534 - `last_heartbeat: string`

35153535

3516 RFC 3339 timestamp of the actual heartbeat from DB

3536 RFC 3339 timestamp of the most recent heartbeat recorded for this work item, or an empty string if none has been recorded

35173537

35183538 - `lease_extended: boolean`

35193539

from line 3715

36953715

36963716 - `"analytics-usage-cost-2026-08-24"`

36973717

3718 - `"experimental-claude-tag-2026-10-21"`

3719

36983720- `"anthropic-workspace-id": optional string`

36993721

37003722 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 3995

39733995

39743996 - `"analytics-usage-cost-2026-08-24"`

39753997

3998 - `"experimental-claude-tag-2026-10-21"`

3999

39764000#### Returns

39774001

39784002- `BetaSelfHostedWorkListResponse object`

from line 4258

42344258

42354259 - `"analytics-usage-cost-2026-08-24"`

42364260

4261 - `"experimental-claude-tag-2026-10-21"`

4262

42374263- `"anthropic-workspace-id": optional string`

42384264

42394265 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 4526

45004526

45014527 - `"analytics-usage-cost-2026-08-24"`

45024528

4529 - `"experimental-claude-tag-2026-10-21"`

4530

45034531- `"anthropic-workspace-id": optional string`

45044532

45054533 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 4540

45124540

45134541 Statistics about the work queue for an environment.

45144542

4515 Uses Redis Stream consumer group metrics for O(1) queries.

4516

45174543 - `type: "work_queue_stats"`

45184544

45194545 The type of object

from line 4548

45224548

45234549 - `depth: number`

45244550

4525 Number of work items waiting to be picked up (lag from consumer group)

4551 Number of work items waiting to be picked up

45264552

45274553 - `oldest_queued_at: string or null`

45284554

api/beta/environments/work Changed · +20 / -8 lines

from line 141

141141

142142 - `"analytics-usage-cost-2026-08-24"`

143143

144 - `"experimental-claude-tag-2026-10-21"`

145

144146- `"anthropic-workspace-id": optional string`

145147

146148 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 413

411413

412414 - `"analytics-usage-cost-2026-08-24"`

413415

416 - `"experimental-claude-tag-2026-10-21"`

417

414418- `"Anthropic-Worker-ID": optional string`

415419

416420 Unique identifier for the specific worker polling, used to track aggregated environment-level work metrics in Console

from line 671

667671

668672 - `"analytics-usage-cost-2026-08-24"`

669673

674 - `"experimental-claude-tag-2026-10-21"`

675

670676### Returns

671677

672678- `BetaSelfHostedWork object`

from line 936

930936

931937 - `"analytics-usage-cost-2026-08-24"`

932938

939 - `"experimental-claude-tag-2026-10-21"`

940

933941### Returns

934942

935943- `BetaSelfHostedWorkHeartbeatResponse object`

from line 952

944952

945953 - `last_heartbeat: string`

946954

947 RFC 3339 timestamp of the actual heartbeat from DB

955 RFC 3339 timestamp of the most recent heartbeat recorded for this work item, or an empty string if none has been recorded

948956

949957 - `lease_extended: boolean`

950958

from line 1134

11261134

11271135 - `"analytics-usage-cost-2026-08-24"`

11281136

1137 - `"experimental-claude-tag-2026-10-21"`

1138

11291139- `"anthropic-workspace-id": optional string`

11301140

11311141 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1414

14041414

14051415 - `"analytics-usage-cost-2026-08-24"`

14061416

1417 - `"experimental-claude-tag-2026-10-21"`

1418

14071419### Returns

14081420

14091421- `BetaSelfHostedWorkListResponse object`

from line 1677

16651677

16661678 - `"analytics-usage-cost-2026-08-24"`

16671679

1680 - `"experimental-claude-tag-2026-10-21"`

1681

16681682- `"anthropic-workspace-id": optional string`

16691683

16701684 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1945

19311945

19321946 - `"analytics-usage-cost-2026-08-24"`

19331947

1948 - `"experimental-claude-tag-2026-10-21"`

1949

19341950- `"anthropic-workspace-id": optional string`

19351951

19361952 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1959

19431959

19441960 Statistics about the work queue for an environment.

19451961

1946 Uses Redis Stream consumer group metrics for O(1) queries.

1947

19481962 - `type: "work_queue_stats"`

19491963

19501964 The type of object

from line 1967

19531967

19541968 - `depth: number`

19551969

1956 Number of work items waiting to be picked up (lag from consumer group)

1970 Number of work items waiting to be picked up

19571971

19581972 - `oldest_queued_at: string or null`

19591973

from line 2102

20882102

20892103 - `last_heartbeat: string`

20902104

2091 RFC 3339 timestamp of the actual heartbeat from DB

2105 RFC 3339 timestamp of the most recent heartbeat recorded for this work item, or an empty string if none has been recorded

20922106

20932107 - `lease_extended: boolean`

20942108

from line 2218

22042218

22052219 Statistics about the work queue for an environment.

22062220

2207 Uses Redis Stream consumer group metrics for O(1) queries.

2208

22092221 - `type: "work_queue_stats"`

22102222

22112223 The type of object

from line 2226

22142226

22152227 - `depth: number`

22162228

2217 Number of work items waiting to be picked up (lag from consumer group)

2229 Number of work items waiting to be picked up

22182230

22192231 - `oldest_queued_at: string or null`

22202232

api/beta/environments/work/stats Changed · +3 / -3 lines

from line 135

135135

136136 - `"analytics-usage-cost-2026-08-24"`

137137

138 - `"experimental-claude-tag-2026-10-21"`

139

138140- `"anthropic-workspace-id": optional string`

139141

140142 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 149

147149

148150 Statistics about the work queue for an environment.

149151

150 Uses Redis Stream consumer group metrics for O(1) queries.

151

152152 - `type: "work_queue_stats"`

153153

154154 The type of object

from line 157

157157

158158 - `depth: number`

159159

160 Number of work items waiting to be picked up (lag from consumer group)

160 Number of work items waiting to be picked up

161161

162162 - `oldest_queued_at: string or null`

163163

api/beta/files Changed · +10 / -0 lines

from line 133

133133

134134 - `"analytics-usage-cost-2026-08-24"`

135135

136 - `"experimental-claude-tag-2026-10-21"`

137

136138- `"anthropic-workspace-id": optional string`

137139

138140 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 398

396398

397399 - `"analytics-usage-cost-2026-08-24"`

398400

401 - `"experimental-claude-tag-2026-10-21"`

402

399403- `"anthropic-workspace-id": optional string`

400404

401405 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 642

638642

639643 - `"analytics-usage-cost-2026-08-24"`

640644

645 - `"experimental-claude-tag-2026-10-21"`

646

641647- `"anthropic-workspace-id": optional string`

642648

643649 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 792

786792

787793 - `"analytics-usage-cost-2026-08-24"`

788794

795 - `"experimental-claude-tag-2026-10-21"`

796

789797- `"anthropic-workspace-id": optional string`

790798

791799 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1024

10161024 - `"message-threads-2026-08-12"`

10171025

10181026 - `"analytics-usage-cost-2026-08-24"`

1027

1028 - `"experimental-claude-tag-2026-10-21"`

10191029

10201030- `"anthropic-workspace-id": optional string`

10211031

api/beta/memory_stores Changed · +28 / -0 lines

from line 133

133133

134134 - `"analytics-usage-cost-2026-08-24"`

135135

136 - `"experimental-claude-tag-2026-10-21"`

137

136138- `"anthropic-workspace-id": optional string`

137139

138140 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 401

399401

400402 - `"analytics-usage-cost-2026-08-24"`

401403

404 - `"experimental-claude-tag-2026-10-21"`

405

402406- `"anthropic-workspace-id": optional string`

403407

404408 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 626

622626

623627 - `"analytics-usage-cost-2026-08-24"`

624628

629 - `"experimental-claude-tag-2026-10-21"`

630

625631- `"anthropic-workspace-id": optional string`

626632

627633 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 846

840846

841847 - `"analytics-usage-cost-2026-08-24"`

842848

849 - `"experimental-claude-tag-2026-10-21"`

850

843851- `"anthropic-workspace-id": optional string`

844852

845853 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1090

10821090

10831091 - `"analytics-usage-cost-2026-08-24"`

10841092

1093 - `"experimental-claude-tag-2026-10-21"`

1094

10851095- `"anthropic-workspace-id": optional string`

10861096

10871097 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1265

12551265

12561266 - `"analytics-usage-cost-2026-08-24"`

12571267

1268 - `"experimental-claude-tag-2026-10-21"`

1269

12581270- `"anthropic-workspace-id": optional string`

12591271

12601272 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1566

15541566

15551567 - `"analytics-usage-cost-2026-08-24"`

15561568

1569 - `"experimental-claude-tag-2026-10-21"`

1570

15571571- `"anthropic-workspace-id": optional string`

15581572

15591573 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1861

18471861

18481862 - `"analytics-usage-cost-2026-08-24"`

18491863

1864 - `"experimental-claude-tag-2026-10-21"`

1865

18501866- `"anthropic-workspace-id": optional string`

18511867

18521868 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2138

21222138

21232139 - `"analytics-usage-cost-2026-08-24"`

21242140

2141 - `"experimental-claude-tag-2026-10-21"`

2142

21252143- `"anthropic-workspace-id": optional string`

21262144

21272145 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2390

23722390

23732391 - `"analytics-usage-cost-2026-08-24"`

23742392

2393 - `"experimental-claude-tag-2026-10-21"`

2394

23752395- `"anthropic-workspace-id": optional string`

23762396

23772397 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2662

26422662

26432663 - `"analytics-usage-cost-2026-08-24"`

26442664

2665 - `"experimental-claude-tag-2026-10-21"`

2666

26452667- `"anthropic-workspace-id": optional string`

26462668

26472669 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 2915

28932915

28942916 - `"analytics-usage-cost-2026-08-24"`

28952917

2918 - `"experimental-claude-tag-2026-10-21"`

2919

28962920- `"anthropic-workspace-id": optional string`

28972921

28982922 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 3253

32293253

32303254 - `"analytics-usage-cost-2026-08-24"`

32313255

3256 - `"experimental-claude-tag-2026-10-21"`

3257

32323258- `"anthropic-workspace-id": optional string`

32333259

32343260 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 3571

35453571 - `"message-threads-2026-08-12"`

35463572

35473573 - `"analytics-usage-cost-2026-08-24"`

3574

3575 - `"experimental-claude-tag-2026-10-21"`

35483576

35493577- `"anthropic-workspace-id": optional string`

35503578

api/beta/memory_stores/memories Changed · +10 / -0 lines

from line 155

155155

156156 - `"analytics-usage-cost-2026-08-24"`

157157

158 - `"experimental-claude-tag-2026-10-21"`

159

158160- `"anthropic-workspace-id": optional string`

159161

160162 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 450

448450

449451 - `"analytics-usage-cost-2026-08-24"`

450452

453 - `"experimental-claude-tag-2026-10-21"`

454

451455- `"anthropic-workspace-id": optional string`

452456

453457 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 727

723727

724728 - `"analytics-usage-cost-2026-08-24"`

725729

730 - `"experimental-claude-tag-2026-10-21"`

731

726732- `"anthropic-workspace-id": optional string`

727733

728734 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 979

973979

974980 - `"analytics-usage-cost-2026-08-24"`

975981

982 - `"experimental-claude-tag-2026-10-21"`

983

976984- `"anthropic-workspace-id": optional string`

977985

978986 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 1250

12421250 - `"message-threads-2026-08-12"`

12431251

12441252 - `"analytics-usage-cost-2026-08-24"`

1253

1254 - `"experimental-claude-tag-2026-10-21"`

12451255

12461256- `"anthropic-workspace-id": optional string`

12471257

api/beta/memory_stores/memory_versions Changed · +6 / -0 lines

from line 213

213213

214214 - `"analytics-usage-cost-2026-08-24"`

215215

216 - `"experimental-claude-tag-2026-10-21"`

217

216218- `"anthropic-workspace-id": optional string`

217219

218220 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 551

549551

550552 - `"analytics-usage-cost-2026-08-24"`

551553

554 - `"experimental-claude-tag-2026-10-21"`

555

552556- `"anthropic-workspace-id": optional string`

553557

554558 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 869

865869 - `"message-threads-2026-08-12"`

866870

867871 - `"analytics-usage-cost-2026-08-24"`

872

873 - `"experimental-claude-tag-2026-10-21"`

868874

869875- `"anthropic-workspace-id": optional string`

870876

api/beta/messages Changed · +218 / -50 lines

This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.

from line 137

137137

138138 - `"analytics-usage-cost-2026-08-24"`

139139

140 - `"experimental-claude-tag-2026-10-21"`

141

140142- `"anthropic-user-profile-id": optional string`

141143

142144 The user profile ID to attribute this request to. Use when acting on behalf of a party other than your organization.

from line 1038

10361038

10371039 - `BetaEncryptedCodeExecutionResultBlockParam object`

10381040

1039 Code execution result with encrypted stdout for PFC + web_search results.

1041 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form, sent back in a later request.

10401042

10411043 - `type: "encrypted_code_execution_result"`

10421044

from line 1574

15721574

15731575 - `BetaCodeExecutionTool20260120 object`

15741576

1575 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

1577 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

15761578

1579 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

1580

1581 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

1582

15771583 - `type: "code_execution_20260120"`

15781584

15791585 - `name: "code_execution"`

from line 4226

42204226

42214227 - `BetaCodeExecutionTool20260120 object`

42224228

4223 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

4229 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

42244230

4231 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

4232

4233 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

4234

42254235 - `BetaCodeExecutionTool20260521 object`

42264236

42274237 Code execution tool with REPL state persistence.

from line 4926

49164926

49174927 - `BetaEncryptedCodeExecutionResultBlock object`

49184928

4919 Code execution result with encrypted stdout for PFC + web_search results.

4929 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form.

49204930

4931 The API returns this block in place of `code_execution_result` when the code called the web search tool.

4932

4933 See the [programmatic tool calling guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling?sdk_languages=http) to learn more about programmatic tool calling.

4934

49214935 - `type: "encrypted_code_execution_result"`

49224936

49234937 default: encrypted_code_execution_result

from line 5492

54785492

54795493 - `BetaCodeExecutionTool20260120 object`

54805494

5481 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

5495 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

54825496

5497 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

5498

5499 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

5500

54835501 - `type: "code_execution_20260120"`

54845502

54855503 - `name: "code_execution"`

from line 7020

70027020

70037021 - `response_inclusion: optional "full" or "excluded"`

70047022

7005 How this tool's result blocks appear in the API response when the result was consumed by a completed code_execution call in the same turn. 'full' returns the complete content (default). 'excluded' drops the nested server_tool_use and result block pair entirely. Results from direct calls, or from code_execution calls that paused before completing, are always returned in full so they can be sent back on the next turn.

7006

7007 - `"full"`

7008

7009 - `"excluded"`

7010

7011 - `strict: optional boolean`

7012

7013 When true, guarantees schema validation on tool names and inputs

7014

7015 - `url_sources: optional BetaWebFetchURLSources or null`

7016

7017 Which sources contribute to the set of URLs the tool may fetch. Omitted means every source.

7018

7019 - `use_cache: optional boolean`

7020

7021 Whether to use cached content. Set to false to bypass the cache and fetch fresh content. Only set to false when the user explicitly requests fresh content or when fetching rapidly-changing sources.

7022

7023 - `BetaAdvisorTool20260301 object`

7024

7025 - `type: "advisor_20260301"`

7026

7027 - `model: Model`

7028

7029 The model that will complete your prompt.

7030

7031 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.

7032

7033 - `"claude-haiku-5-5"`

7034

7035 Fastest model for high-volume, real-time tasks

7036

7037 - `"claude-sonnet-5-5"`

7038

7039 Efficient model for coding and agents

7040

7041 - `"claude-fable-5-1"`

7042

7043 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows

7044

7045 - `"claude-opus-5-5"`

7046

7047 Powerful intelligence for coding, knowledge work, and long-running agents

7048

7049

7023 How this tool's result blocks appear in the

api/beta/messages/batches Changed · +63 / -11 lines

This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.

from line 137

137137

138138 - `"analytics-usage-cost-2026-08-24"`

139139

140 - `"experimental-claude-tag-2026-10-21"`

141

140142- `"anthropic-user-profile-id": optional string`

141143

142144 The user profile ID to attribute the requests in this batch to. Use when acting on behalf of a party other than your organization. Applies to every request in the batch; an individual request whose `user_profile_id` body field conflicts with this header is errored.

from line 1058

10561058

10571059 - `BetaEncryptedCodeExecutionResultBlockParam object`

10581060

1059 Code execution result with encrypted stdout for PFC + web_search results.

1061 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form, sent back in a later request.

10601062

10611063 - `type: "encrypted_code_execution_result"`

10621064

from line 1594

15921594

15931595 - `BetaCodeExecutionTool20260120 object`

15941596

1595 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

1597 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

15961598

1599 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

1600

1601 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

1602

15971603 - `type: "code_execution_20260120"`

15981604

15991605 - `name: "code_execution"`

from line 4246

42404246

42414247 - `BetaCodeExecutionTool20260120 object`

42424248

4243 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

4249 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

42444250

4251 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

4252

4253 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

4254

42454255 - `BetaCodeExecutionTool20260521 object`

42464256

42474257 Code execution tool with REPL state persistence.

from line 4623

46134623

46144624 - `"analytics-usage-cost-2026-08-24"`

46154625

4626 - `"experimental-claude-tag-2026-10-21"`

4627

46164628- `"anthropic-workspace-id": optional string`

46174629

46184630 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 4921

49094921

49104922 - `"analytics-usage-cost-2026-08-24"`

49114923

4924 - `"experimental-claude-tag-2026-10-21"`

4925

49124926- `"anthropic-workspace-id": optional string`

49134927

49144928 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 5228

52145228

52155229 - `"analytics-usage-cost-2026-08-24"`

52165230

5231 - `"experimental-claude-tag-2026-10-21"`

5232

52175233- `"anthropic-workspace-id": optional string`

52185234

52195235 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 5517

55015517

55025518 - `"analytics-usage-cost-2026-08-24"`

55035519

5520 - `"experimental-claude-tag-2026-10-21"`

5521

55045522- `"anthropic-workspace-id": optional string`

55055523

55065524 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 5698

56805698

56815699 - `"analytics-usage-cost-2026-08-24"`

56825700

5701 - `"experimental-claude-tag-2026-10-21"`

5702

56835703- `"anthropic-workspace-id": optional string`

56845704

56855705 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 6347

63276347

63286348 - `BetaEncryptedCodeExecutionResultBlock object`

63296349

6330 Code execution result with encrypted stdout for PFC + web_search results.

6350 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form.

63316351

6352 The API returns this block in place of `code_execution_result` when the code called the web search tool.

6353

6354 See the [programmatic tool calling guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling?sdk_languages=http) to learn more about programmatic tool calling.

6355

63326356 - `type: "encrypted_code_execution_result"`

63336357

63346358 default: encrypted_code_execution_result

from line 6913

68896913

68906914 - `BetaCodeExecutionTool20260120 object`

68916915

6892 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

6916 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

68936917

6918 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

6919

6920 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

6921

68946922 - `type: "code_execution_20260120"`

68956923

68966924 - `name: "code_execution"`

from line 7232

72047232

72057233 `new_tab`'s config overrides.

72067234

7207 - `defer_loading: optional boolean or null`

7208

7209 Defer loading for this member. Must resolve to the same value on every enabled member of the toolset.

7210

7211 - `enabled: optional boolean or null`

7212

7213 Whether this member is offered to the model. Default is per member, per the toolset's documentation. A member whose enabled resolves false is withheld from the served schema.

7214

7215 - `read_console: optional BetaBrowserReadConsoleConfig or null`

7216

7217 `read_console`'s config overrides.

7218

7219 - `defer_loading: optional boolean or null`

7220

7221 Defer loading for this member. Must resolve to the same value on every enabled member of the toolset.

7222

7223 - `enabled: optional boolean or null`

7224

7225 Whether this member is offered to the model. Default is per member, per the toolset's documentation. A member whose enabled resolves false is withheld from the served schema.

7226

7227 - `read_network: optional BetaBrowserReadNetworkConfig or null`

7228

7229 `read_network`'s config overrides.

7230

7231 - `defer_loading: optional boolean or null`

7232

7233 Defer loading for this member. Must resolve to the same value on every enabled member of the toolset.

7234

7235 - `enabled: optional boolean or null`

7236

7237 Whether this member is offered to the model. Default is per member, per the toolset's documentation. A member whose enabled resolves false is withheld from the served schema.

7238

7239 - `read_page: optional BetaBrowserReadPageConfig or null`

7240

7241 `read_page`'s config overrides.

7242

7243 - `defer_loading: optional boolean or null`

7244

7245 Defer loading for this member. Must resolve to the same value on every enabled member of the toolset.

7246

7247 - `enabled: optional boolean or null`

7248

7249 Whether this member i

7235 - `defer_loading: optiona

api/beta/messages/batches/create Changed · +13 / -3 lines

from line 135

135135

136136 - `"analytics-usage-cost-2026-08-24"`

137137

138 - `"experimental-claude-tag-2026-10-21"`

139

138140- `"anthropic-user-profile-id": optional string`

139141

140142 The user profile ID to attribute the requests in this batch to. Use when acting on behalf of a party other than your organization. Applies to every request in the batch; an individual request whose `user_profile_id` body field conflicts with this header is errored.

from line 1056

10541056

10551057 - `BetaEncryptedCodeExecutionResultBlockParam object`

10561058

1057 Code execution result with encrypted stdout for PFC + web_search results.

1059 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form, sent back in a later request.

10581060

10591061 - `type: "encrypted_code_execution_result"`

10601062

from line 1592

15901592

15911593 - `BetaCodeExecutionTool20260120 object`

15921594

1593 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

1595 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

15941596

1597 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

1598

1599 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

1600

15951601 - `type: "code_execution_20260120"`

15961602

15971603 - `name: "code_execution"`

from line 4244

42384244

42394245 - `BetaCodeExecutionTool20260120 object`

42404246

4241 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

4247 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

4248

4249 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

4250

4251 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

42424252

42434253 - `BetaCodeExecutionTool20260521 object`

42444254

api/beta/messages/batches/results Changed · +12 / -2 lines

from line 141

141141

142142 - `"analytics-usage-cost-2026-08-24"`

143143

144 - `"experimental-claude-tag-2026-10-21"`

145

144146- `"anthropic-workspace-id": optional string`

145147

146148 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).

from line 790

788790

789791 - `BetaEncryptedCodeExecutionResultBlock object`

790792

791 Code execution result with encrypted stdout for PFC + web_search results.

793 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form.

792794

795 The API returns this block in place of `code_execution_result` when the code called the web search tool.

796

797 See the [programmatic tool calling guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling?sdk_languages=http) to learn more about programmatic tool calling.

798

793799 - `type: "encrypted_code_execution_result"`

794800

795801 default: encrypted_code_execution_result

from line 1356

13501356

13511357 - `BetaCodeExecutionTool20260120 object`

13521358

1353 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

1359 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

1360

1361 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

1362

1363 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

13541364

13551365 - `type: "code_execution_20260120"`

13561366

api/beta/messages/count_tokens Changed · +13 / -3 lines

from line 135

135135

136136 - `"analytics-usage-cost-2026-08-24"`

137137

138 - `"experimental-claude-tag-2026-10-21"`

139

138140- `"anthropic-user-profile-id": optional string`

139141

140142 The user profile ID to attribute this request to. Use when acting on behalf of a party other than your organization.

from line 1024

10221024

10231025 - `BetaEncryptedCodeExecutionResultBlockParam object`

10241026

1025 Code execution result with encrypted stdout for PFC + web_search results.

1027 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form, sent back in a later request.

10261028

10271029 - `type: "encrypted_code_execution_result"`

10281030

from line 1560

15581560

15591561 - `BetaCodeExecutionTool20260120 object`

15601562

1561 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

1563 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

15621564

1565 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

1566

1567 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

1568

15631569 - `type: "code_execution_20260120"`

15641570

15651571 - `name: "code_execution"`

from line 3993

39873993

39883994 - `BetaCodeExecutionTool20260120 object`

39893995

3990 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

3996 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

3997

3998 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

3999

4000 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

39914001

39924002 - `BetaCodeExecutionTool20260521 object`

39934003

api/beta/messages/create Changed · +23 / -5 lines

This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.

from line 135

135135

136136 - `"analytics-usage-cost-2026-08-24"`

137137

138 - `"experimental-claude-tag-2026-10-21"`

139

138140- `"anthropic-user-profile-id": optional string`

139141

140142 The user profile ID to attribute this request to. Use when acting on behalf of a party other than your organization.

from line 1036

10341036

10351037 - `BetaEncryptedCodeExecutionResultBlockParam object`

10361038

1037 Code execution result with encrypted stdout for PFC + web_search results.

1039 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form, sent back in a later request.

10381040

10391041 - `type: "encrypted_code_execution_result"`

10401042

from line 1572

15701572

15711573 - `BetaCodeExecutionTool20260120 object`

15721574

1573 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

1575 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

15741576

1577 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

1578

1579 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

1580

15751581 - `type: "code_execution_20260120"`

15761582

15771583 - `name: "code_execution"`

from line 4224

42184224

42194225 - `BetaCodeExecutionTool20260120 object`

42204226

4221 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

4227 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

42224228

4229 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

4230

4231 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

4232

42234233 - `BetaCodeExecutionTool20260521 object`

42244234

42254235 Code execution tool with REPL state persistence.

from line 4924

49144924

49154925 - `BetaEncryptedCodeExecutionResultBlock object`

49164926

4917 Code execution result with encrypted stdout for PFC + web_search results.

4927 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form.

49184928

4929 The API returns this block in place of `code_execution_result` when the code called the web search tool.

4930

4931 See the [programmatic tool calling guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling?sdk_languages=http) to learn more about programmatic tool calling.

4932

49194933 - `type: "encrypted_code_execution_result"`

49204934

49214935 default: encrypted_code_execution_result

from line 5490

54765490

54775491 - `BetaCodeExecutionTool20260120 object`

54785492

5479 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).

5493 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.

54805494

5495 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.

5496

5497 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.

5498

54815499 - `type: "code_execution_20260120"`

54825500

54835501 - `name: "code_execution"`

from line 7018

70007018

70017019 - `response_inclusion: optional "full" or "excluded"`

70027020

7003 How this tool's result blocks appear in the API response when the result was consumed by a completed code_execution call in the same turn. 'full' returns the complete content (default). 'excluded' drops the nested server_tool_use and result block pair entirely. Results from direct calls, or from code_execution calls that paused before completing, are always returned in full so they can be sent back on the next turn.

7004

7005 - `"full"`

7006

7007 - `"excluded"`

7008

7009 - `strict: optional boolean`

7010

7011 When true, guarantees schema validation on tool names and inputs

7012

7013 - `url_sources: optional BetaWebFetchURLSources or null`

7014

7015 Which sources contribute to the set of URLs the tool may fetch. Omitted means every source.

7016

7017 - `use_cache: optional boolean`

7018

7019 Whether to use cached content. Set to false to bypass the cache and fetch fresh content. Only set to false when the user explicitly requests fresh content or when fetching rapidly-changing sources.

7020

7021 - `BetaAdvisorTool20260301 object`

7022

7023 - `type: "advisor_20260301"`

7024

7025 - `model: Model`

7026

7027 The model that will complete your prompt.

7028

7029 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.

7030

7031 - `"claude-haiku-5-5"`

7032

7033 Fastest model for high-volume, real-time tasks

7034

7035 - `"claude-sonnet-5-5"`

7036

7037 Efficient model for coding and agents

7038

7039 - `"claude-fable-5-1"`

7040

7041 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows

7042

7043 - `"claude-opus-5-5"`

7044

7045 Powerful intelligence for coding, knowledge work, and long-running agents

7046

7047

7021 How this tool's result blocks appear in the

api/beta/organization Changed · +4688 / -204 lines

## Organization › Claude Tag › Slack Channels ### Set Up Slack Channel ### List Slack Channels ### Get Slack Channel ### Update Slack Channel ### Remove Slack Channel ## Organization › Claude Tag › Slack Workspaces ### List Slack Workspaces ### Get Slack Workspace ### Update Slack Workspace ## Organization › Claude Tag › Bundles ### List Bundles ### Get Bundle ## Organization › Claude Tag › Bundles › Targets ### List Bundle Targets ### Attach Bundle ### Get Bundle Target ### Detach Bundle

This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.

The two sides of this change are more than 400 edits apart, too far apart to line up, so this is the differ's own diff of it and the words inside a line are not marked.

from line 306

306306 "id": "apikey_01Rj2N8SVvo6BePZj99NhmiT",

307307 "created_at": "2024-10-30T23:58:27.427722Z",

308308 "created_by": {

309 "id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

309 "id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

310310 "type": "user"

311311 },

312312 "expires_at": "2024-10-30T23:58:27.427722Z",

from line 314

314314 "partial_key_hint": "sk-ant-api03-R2D...igAA",

315315 "principal": {

316316 "type": "user_actor",

317 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

317 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

318318 },

319319 "scope": {

320320 "type": "workspace",

from line 483

483483 "id": "apikey_01Rj2N8SVvo6BePZj99NhmiT",

484484 "created_at": "2024-10-30T23:58:27.427722Z",

485485 "created_by": {

486 "id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

486 "id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

487487 "type": "user"

488488 },

489489 "expires_at": "2024-10-30T23:58:27.427722Z",

from line 491

491491 "partial_key_hint": "sk-ant-api03-R2D...igAA",

492492 "principal": {

493493 "type": "user_actor",

494 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

494 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

495495 },

496496 "scope": {

497497 "type": "workspace",

from line 675

675675 "id": "apikey_01Rj2N8SVvo6BePZj99NhmiT",

676676 "created_at": "2024-10-30T23:58:27.427722Z",

677677 "created_by": {

678 "id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

678 "id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

679679 "type": "user"

680680 },

681681 "expires_at": "2024-10-30T23:58:27.427722Z",

from line 683

683683 "partial_key_hint": "sk-ant-api03-R2D...igAA",

684684 "principal": {

685685 "type": "user_actor",

686 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

686 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

687687 },

688688 "scope": {

689689 "type": "workspace",

from line 722

722722

723723 - `kms_arn: string`

724724

725 Full ARN of the AWS KMS key. On Claude Platform on AWS the key must be a single-Region key in your organization's own AWS account; cross-account keys, multi-Region keys, and alias ARNs are rejected.

725 Full ARN of the AWS KMS key. On Claude Platform on AWS, alias ARNs are rejected, and your organization's own key must be a single-Region key in your organization's AWS account.

726726

727727 maxLength: 2048

728728

from line 734

734734

735735 **Deprecated**

736736

737 IAM role ARN. Deprecated — Anthropic reaches the KMS key through its own intermediate role (or, on Claude Platform on AWS, with credentials AWS issues for the Workspace); this field is ignored.

737 IAM role ARN. Deprecated; this field is ignored.

738738

739739 - `BetaGCPExternalKeyConfig object`

740740

from line 776

776776

777777 Data residency geo. Only `us` is supported.

778778

779- `user_profile_id: optional string or null`

780

781 ID of a user profile to bind the key to. Omit this field for a key that isn't bound to a user profile. Such a key can be attached to a Workspace whatever its user profile. The binding cannot be changed later. A key bound to a user profile can only be attached to a Workspace that has the same user profile.

782

779783#### Returns

780784

781785- `BetaExternalKey object`

from line 796

792796

793797 - `id: string`

794798

795 Identifier of the external key config. A tagged ID prefixed `ekey_`, or — for organizations on the Claude Platform on AWS — the AWS KMS key ARN.

799 Identifier of the external key config: a tagged ID prefixed `ekey_`, or, on Claude Platform on AWS, an AWS KMS key's ARN.

796800

797801 - `attachment: BetaExternalKeyAttachedAttachment or BetaExternalKeyUnattachedAttachment`

798802

from line 824

820824

821825 - `geo: string`

822826

823 Data residency geo. Selects which regional validator handles this key's encrypt/decrypt roundtrips.

827 Data residency geo. Selects the region that handles this key's encrypt/decrypt roundtrips.

824828

825829 - `provider_config: BetaAWSExternalKeyConfig or BetaGCPExternalKeyConfig or BetaAzureExternalKeyConfig`

826830

from line 836

832836

833837 - `kms_arn: string`

834838

835 Full ARN of the AWS KMS key. On Claude Platform on AWS the key must be a single-Region key in your organization's own AWS account; cross-account keys, multi-Region keys, and alias ARNs are rejected.

839 Full ARN of the AWS KMS key. On Claude Platform on AWS, alias ARNs are rejected, and your organization's own key must be a single-Region key in your organization's AWS account.

836840

837841 maxLength: 2048

838842

from line 848

844848

845849 **Deprecated**

846850

847 IAM role ARN. Deprecated — Anthropic reaches the KMS key through its own intermediate role (or, on Claude Platform on AWS, with credentials AWS issues for the Workspace); this field is ignored.

851 IAM role ARN. Deprecated; this field is ignored.

848852

849853 - `BetaGCPExternalKeyConfig object`

850854

from line 882

878882

879883 format: date-time

880884

885 - `user_profile_id: string or null`

886

887 ID of the user profile the key is bound to, or null if the key is not bound to one. A key bound to a user profile can only be attached to a Workspace that has the same user profile. A key that isn't bound can be attached to a Workspace whatever its user profile.

888

881889#### Example

882890

883891```bash

from line 897

889897 "provider_config": {

890898 "kms_arn": "arn:aws:kms:us-east-1:111122223333:key/abcd1234-5678-90ab-cdef-000011112222",

891899 "type": "aws"

892 }

900 },

901 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

893902 }'

894903```

895904

from line 920

911920 "role_arn": "arn:aws:iam::111122223333:role/anthropic-cmek"

912921 },

913922 "type": "external_key",

914 "updated_at": "2024-10-30T23:58:27.427722Z"

923 "updated_at": "2024-10-30T23:58:27.427722Z",

924 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

915925}

916926```

917927

from line 964

954964

955965 - `id: string`

956966

957 Identifier of the external key config. A tagged ID prefixed `ekey_`, or — for organizations on the Claude Platform on AWS — the AWS KMS key ARN.

967 Identifier of the external key config: a tagged ID prefixed `ekey_`, or, on Claude Platform on AWS, an AWS KMS key's ARN.

958968

959969 - `attachment: BetaExternalKeyAttachedAttachment or BetaExternalKeyUnattachedAttachment`

960970

from line 992

982992

983993 - `geo: string`

984994

985 Data residency geo. Selects which regional validator handles this key's encrypt/decrypt roundtrips.

995 Data residency geo. Selects the region that handles this key's encrypt/decrypt roundtrips.

986996

987997 - `provider_config: BetaAWSExternalKeyConfig or BetaGCPExternalKeyConfig or BetaAzureExternalKeyConfig`

988998

from line 1004

9941004

9951005 - `kms_arn: string`

9961006

997 Full ARN of the AWS KMS key. On Claude Platform on AWS the key must be a single-Region key in your organization's own AWS account; cross-account keys, multi-Region keys, and alias ARNs are rejected.

1007 Full ARN of the AWS KMS key. On Claude Platform on AWS, alias ARNs are rejected, and your organization's own key must be a single-Region key in your organization's AWS account.

9981008

9991009 maxLength: 2048

10001010

from line 1016

10061016

10071017 **Deprecated**

10081018

1009 IAM role ARN. Deprecated — Anthropic reaches the KMS key through its own intermediate role (or, on Claude Platform on AWS, with credentials AWS issues for the Workspace); this field is ignored.

1019 IAM role ARN. Deprecated; this field is ignored.

10101020

10111021 - `BetaGCPExternalKeyConfig object`

10121022

from line 1050

10401050

10411051 format: date-time

10421052

1053 - `user_profile_id: string or null`

1054

1055 ID of the user profile the key is bound to, or null if the key is not bound to one. A key bound to a user profile can only be attached to a Workspace that has the same user profile. A key that isn't bound can be attached to a Workspace whatever its user profile.

1056

10431057- `next_page: string or null`

10441058

10451059 Opaque cursor for the next page, or null if no more results. Pass as `?page=` to fetch the next page.

from line 1086

10721086 "role_arn": "arn:aws:iam::111122223333:role/anthropic-cmek"

10731087 },

10741088 "type": "external_key",

1075 "updated_at": "2024-10-30T23:58:27.427722Z"

1089 "updated_at": "2024-10-30T23:58:27.427722Z",

1090 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

10761091 }

10771092 ],

10781093 "next_page": "next_page"

from line 1132

11171132

11181133 - `id: string`

11191134

1120 Identifier of the external key config. A tagged ID prefixed `ekey_`, or — for organizations on the Claude Platform on AWS — the AWS KMS key ARN.

1135 Identifier of the external key config: a tagged ID prefixed `ekey_`, or, on Claude Platform on AWS, an AWS KMS key's ARN.

11211136

11221137 - `attachment: BetaExternalKeyAttachedAttachment or BetaExternalKeyUnattachedAttachment`

11231138

from line 1160

11451160

11461161 - `geo: string`

11471162

1148 Data residency geo. Selects which regional validator handles this key's encrypt/decrypt roundtrips.

1163 Data residency geo. Selects the region that handles this key's encrypt/decrypt roundtrips.

11491164

11501165 - `provider_config: BetaAWSExternalKeyConfig or BetaGCPExternalKeyConfig or BetaAzureExternalKeyConfig`

11511166

from line 1172

11571172

11581173 - `kms_arn: string`

11591174

1160 Full ARN of the AWS KMS key. On Claude Platform on AWS the key must be a single-Region key in your organization's own AWS account; cross-account keys, multi-Region keys, and alias ARNs are rejected.

1175 Full ARN of the AWS KMS key. On Claude Platform on AWS, alias ARNs are rejected, and your organization's own key must be a single-Region key in your organization's AWS account.

11611176

11621177 maxLength: 2048

11631178

from line 1184

11691184

11701185 **Deprecated**

11711186

1172 IAM role ARN. Deprecated — Anthropic reaches the KMS key through its own intermediate role (or, on Claude Platform on AWS, with credentials AWS issues for the Workspace); this field is ignored.

1187 IAM role ARN. Deprecated; this field is ignored.

11731188

11741189 - `BetaGCPExternalKeyConfig object`

11751190

from line 1218

12031218

12041219 format: date-time

12051220

1221 - `user_profile_id: string or null`

1222

1223 ID of the user profile the key is bound to, or null if the key is not bound to one. A key bound to a user profile can only be attached to a Workspace that has the same user profile. A key that isn't bound can be attached to a Workspace whatever its user profile.

1224

12061225#### Example

12071226

12081227```bash

from line 1248

12291248 "role_arn": "arn:aws:iam::111122223333:role/anthropic-cmek"

12301249 },

12311250 "type": "external_key",

1232 "updated_at": "2024-10-30T23:58:27.427722Z"

1251 "updated_at": "2024-10-30T23:58:27.427722Z",

1252 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

12331253}

12341254```

12351255

from line 1301

12811301

12821302 - `kms_arn: string`

12831303

1284 Full ARN of the AWS KMS key. On Claude Platform on AWS the key must be a single-Region key in your organization's own AWS account; cross-account keys, multi-Region keys, and alias ARNs are rejected.

1304 Full ARN of the AWS KMS key. On Claude Platform on AWS, alias ARNs are rejected, and your organization's own key must be a single-Region key in your organization's AWS account.

12851305

12861306 maxLength: 2048

12871307

from line 1313

12931313

12941314 **Deprecated**

12951315

1296 IAM role ARN. Deprecated — Anthropic reaches the KMS key through its own intermediate role (or, on Claude Platform on AWS, with credentials AWS issues for the Workspace); this field is ignored.

1316 IAM role ARN. Deprecated; this field is ignored.

12971317

12981318 - `BetaGCPExternalKeyConfig object`

12991319

from line 1361

13411361

13421362 - `id: string`

13431363

1344 Identifier of the external key config. A tagged ID prefixed `ekey_`, or — for organizations on the Claude Platform on AWS — the AWS KMS key ARN.

1364 Identifier of the external key config: a tagged ID prefixed `ekey_`, or, on Claude Platform on AWS, an AWS KMS key's ARN.

13451365

13461366 - `attachment: BetaExternalKeyAttachedAttachment or BetaExternalKeyUnattachedAttachment`

13471367

from line 1389

13691389

13701390 - `geo: string`

13711391

1372 Data residency geo. Selects which regional validator handles this key's encrypt/decrypt roundtrips.

1392 Data residency geo. Selects the region that handles this key's encrypt/decrypt roundtrips.

13731393

13741394 - `provider_config: BetaAWSExternalKeyConfig or BetaGCPExternalKeyConfig or BetaAzureExternalKeyConfig`

13751395

from line 1401

13811401

13821402 - `kms_arn: string`

13831403

1384 Full ARN of the AWS KMS key. On Claude Platform on AWS the key must be a single-Region key in your organization's own AWS account; cross-account keys, multi-Region keys, and alias ARNs are rejected.

1404 Full ARN of the AWS KMS key. On Claude Platform on AWS, alias ARNs are rejected, and your organization's own key must be a single-Region key in your organization's AWS account.

13851405

13861406 maxLength: 2048

13871407

from line 1413

13931413

13941414 **Deprecated**

13951415

1396 IAM role ARN. Deprecated — Anthropic reaches the KMS key through its own intermediate role (or, on Claude Platform on AWS, with credentials AWS issues for the Workspace); this field is ignored.

1416 IAM role ARN. Deprecated; this field is ignored.

13971417

13981418 - `BetaGCPExternalKeyConfig object`

13991419

from line 1447

14271447

14281448 format: date-time

14291449

1450 - `user_profile_id: string or null`

1451

1452 ID of the user profile the key is bound to, or null if the key is not bound to one. A key bound to a user profile can only be attached to a Workspace that has the same user profile. A key that isn't bound can be attached to a Workspace whatever its user profile.

1453

14301454#### Example

14311455

14321456```bash

from line 1479

14551479 "role_arn": "arn:aws:iam::111122223333:role/anthropic-cmek"

14561480 },

14571481 "type": "external_key",

1458 "updated_at": "2024-10-30T23:58:27.427722Z"

1482 "updated_at": "2024-10-30T23:58:27.427722Z",

1483 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

14591484}

14601485```

14611486

from line 1742

17171742

17181743 - `"analytics-usage-cost-2026-08-24"`

17191744

1745 - `"experimental-claude-tag-2026-10-21"`

1746

17201747#### Body parameters

17211748

17221749- `issuer_url: string`

from line 2150

21232150

21242151 - `"analytics-usage-cost-2026-08-24"`

21252152

2153 - `"experimental-claude-tag-2026-10-21"`

2154

21262155#### Returns

21272156

21282157- `data: array of BetaFederationIssuer`

from line 2469

24402469

24412470 - `"analytics-usage-cost-2026-08-24"`

24422471

2472 - `"experimental-claude-tag-2026-10-21"`

2473

24432474#### Returns

24442475

24452476- `BetaFederationIssuer object`

from line 2791

27602791

27612792 - `"analytics-usage-cost-2026-08-24"`

27622793

2794 - `"experimental-claude-tag-2026-10-21"`

2795

27632796#### Body parameters

27642797

27652798- `check_jti: optional boolean or null`

from line 3191

31583191

31593192 - `"analytics-usage-cost-2026-08-24"`

31603193

3194 - `"experimental-claude-tag-2026-10-21"`

3195

31613196#### Returns

31623197

31633198- `BetaFederationIssuer object`

from line 3517

34823517

34833518 - `"analytics-usage-cost-2026-08-24"`

34843519

3520 - `"experimental-claude-tag-2026-10-21"`

3521

34853522#### Body parameters

34863523

34873524- `issuer_id: string`

from line 3950

39133950

39143951 - `"analytics-usage-cost-2026-08-24"`

39153952

3953 - `"experimental-claude-tag-2026-10-21"`

3954

39163955#### Returns

39173956

39183957- `data: array of BetaFederationRule`

from line 4269

42304269

42314270 - `"analytics-usage-cost-2026-08-24"`

42324271

4272 - `"experimental-claude-tag-2026-10-21"`

4273

42334274#### Returns

42344275

42354276- `BetaFederationRule object`

from line 4606

45654606

45664607 - `"analytics-usage-cost-2026-08-24"`

45674608

4609 - `"experimental-claude-tag-2026-10-21"`

4610

45684611#### Body parameters

45694612

45704613- `applies_to_all_workspaces: optional boolean or null`

from line 5015

49725015

49735016 - `"analytics-usage-cost-2026-08-24"`

49745017

5018 - `"experimental-claude-tag-2026-10-21"`

5019

49755020#### Returns

49765021

49775022- `BetaFederationRule object`

from line 5348

53035348

53045349 - `"analytics-usage-cost-2026-08-24"`

53055350

5351 - `"experimental-claude-tag-2026-10-21"`

5352

53065353#### Body parameters

53075354

53085355- `workspace_id: string`

from line 5565

55185565

55195566 - `"analytics-usage-cost-2026-08-24"`

55205567

5568 - `"experimental-claude-tag-2026-10-21"`

5569

55215570#### Returns

55225571

55235572- `data: array of BetaFederationRuleWorkspace`

from line 5772

57235772

57245773 - `"analytics-usage-cost-2026-08-24"`

57255774

5775 - `"experimental-claude-tag-2026-10-21"`

5776

57265777#### Returns

57275778

57285779- `type: "federation_rule_workspace_deleted"`

from line 6451

64006451

64016452 - `"analytics-usage-cost-2026-08-24"`

64026453

6454 - `"experimental-claude-tag-2026-10-21"`

6455

64036456#### Body parameters

64046457

64056458- `name: string`

from line 6720

66676720

66686721 - `"analytics-usage-cost-2026-08-24"`

66696722

6723 - `"experimental-claude-tag-2026-10-21"`

6724

66706725#### Returns

66716726

66726727- `data: array of BetaServiceAccount`

from line 6951

68966951

68976952 - `"analytics-usage-cost-2026-08-24"`

68986953

6954 - `"experimental-claude-tag-2026-10-21"`

6955

68996956#### Returns

69006957

69016958- `BetaServiceAccount object`

from line 7183

71267183

71277184 - `"analytics-usage-cost-2026-08-24"`

71287185

7186 - `"experimental-claude-tag-2026-10-21"`

7187

71297188#### Body parameters

71307189

71317190- `description: optional string or null`

from line 7433

73747433

73757434 - `"analytics-usage-cost-2026-08-24"`

73767435

7436 - `"experimental-claude-tag-2026-10-21"`

7437

73777438#### Returns

73787439

73797440- `BetaServiceAccount object`

from line 7670

76097670

76107671 - `"analytics-usage-cost-2026-08-24"`

76117672

7673 - `"experimental-claude-tag-2026-10-21"`

7674

76127675#### Body parameters

76137676

76147677- `workspace_id: string`

from line 7916

78537916

78547917 - `"analytics-usage-cost-2026-08-24"`

78557918

7919 - `"experimental-claude-tag-2026-10-21"`

7920

78567921#### Returns

78577922

78587923- `data: array of BetaServiceAccountWorkspaceMember`

from line 8134

80698134

80708135 - `"analytics-usage-cost-2026-08-24"`

80718136

8137 - `"experimental-claude-tag-2026-10-21"`

8138

80728139#### Returns

80738140

80748141- `type: "service_account_workspace_member_deleted"`

from line 8293

82268293{

82278294 "data": [

82288295 {

8229 "id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

8296 "id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

82308297 "added_at": "2024-10-30T23:58:27.427722Z",

82318298 "email": "[email protected]",

82328299 "name": "Jane Doe",

from line 8391

83248391

83258392```json

83268393{

8327 "id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

8394 "id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

83288395 "added_at": "2024-10-30T23:58:27.427722Z",

83298396 "email": "[email protected]",

83308397 "name": "Jane Doe",

from line 8506

84398506

84408507```json

84418508{

8442 "id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

8509 "id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

84438510 "added_at": "2024-10-30T23:58:27.427722Z",

84448511 "email": "[email protected]",

84458512 "name": "Jane Doe",

from line 8562

84958562

84968563```json

84978564{

8498 "id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

8565 "id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

84998566 "type": "user_deleted"

85008567}

85018568```

from line 8637

85708637

85718638 - `compartment_id: string`

85728639

8573 Identifier for this Workspace's encryption compartment. When you configure a

8574 customer-managed encryption key (CMEK) on AWS, reference this value in your

8575 KMS key-policy condition so the key is scoped to this compartment. On GCP and

8576 Azure, Anthropic enforces the compartment binding automatically; you do not

8577 need to reference this value in your key configuration. See the CMEK

8578 integration guide for the required key configuration; unless your organization

8579 is on Claude Platform on AWS, it includes a separate value used during key

8580 validation. On Claude Platform on AWS there is no separate validation value:

8581 the key is validated against this Workspace's own value when it is attached, so

8582 if your key policy uses the compartment condition, add this value to it before

8583 attaching the key.

8640 Identifier for this Workspace's encryption compartment. When you configure an

8641 AWS KMS customer-managed encryption key (CMEK) that isn't bound to a user

8642 profile, reference this value in your KMS key-policy condition so the key is

8643 scoped to this compartment. On GCP and Azure, Anthropic enforces the compartment

8644 binding automatically; you do not need to reference this value in your key

8645 configuration. See the CMEK integration guide for the required key

8646 configuration; unless your organization is on Claude Platform on AWS, it

8647 includes a separate value used during key validation. On Claude Platform on AWS

8648 there is no separate validation value: your organization's own key is validated

8649 against this Workspace's own value when it is attached, so if its key policy

8650 uses the compartment condition, add this value to it before attaching the key.

85848651

85858652 - `created_at: string`

85868653

from line 8693

86268693 Workspace. Setting this field requires CMEK to be enabled for your

86278694 organization. When set, data stored for this Workspace is encrypted with the

86288695 referenced key. Create key configurations with the External Keys API. On

8629 Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a

8630 single-Region key in the same AWS account and Region as the Workspace. On that

8631 platform the key is validated against this Workspace when it is attached, so a

8632 key-policy problem is reported as an error on this request. This field is write-once:

8633 once a key is attached to a Workspace it cannot be detached or replaced. To

8634 rotate key material, rotate the underlying key on your cloud KMS; the

8635 `external_key_id` stays the same.

8696 Claude Platform on AWS the value for an AWS KMS key is the key's ARN, and your

8697 organization's own key must be a single-Region key in the same AWS account and

8698 Region as the Workspace. On that platform the key is validated against this

8699 Workspace when it is attached, so a key-policy problem is reported as an error

8700 on this request. This field is write-once: once a key is attached to a Workspace

8701 it cannot be detached or replaced. To rotate key material, rotate the underlying

8702 key on your cloud KMS; the `external_key_id` stays the same.

86368703

86378704 - `name: string`

86388705

from line 8709

86428709

86438710 User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.

86448711

8712 - `user_profile_id: string or null`

8713

8714 ID of the user profile attached to the Workspace, or null if it has none. Every inference request through a Workspace that has a user profile must name that profile.

8715

86458716- `first_id: string or null`

86468717

86478718 First ID in the `data` list. Can be used as the `before_id` for the previous page.

from line 8755

86848755 "env": "prod",

86858756 "team": "platform"

86868757 },

8687 "type": "workspace"

8758 "type": "workspace",

8759 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

86888760 }

86898761 ],

86908762 "first_id": "first_id",

from line 8893

88218893

88228894 - `"analytics-usage-cost-2026-08-24"`

88238895

8896 - `"experimental-claude-tag-2026-10-21"`

8897

88248898#### Body parameters

88258899

88268900- `name: string`

from line 8943

88698943 Workspace. Setting this field requires CMEK to be enabled for your

88708944 organization. When set, data stored for this Workspace is encrypted with the

88718945 referenced key. Create key configurations with the External Keys API. On

8872 Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a

8873 single-Region key in the same AWS account and Region as the Workspace. On that

8874 platform the key is validated against this Workspace when it is attached, so a

8875 key-policy problem is reported as an error on this request. This field is write-once:

8876 once a key is attached to a Workspace it cannot be detached or replaced. To

8877 rotate key material, rotate the underlying key on your cloud KMS; the

8878 `external_key_id` stays the same.

8946 Claude Platform on AWS the value for an AWS KMS key is the key's ARN, and your

8947 organization's own key must be a single-Region key in the same AWS account and

8948 Region as the Workspace. On that platform the key is validated against this

8949 Workspace when it is attached, so a key-policy problem is reported as an error

8950 on this request. This field is write-once: once a key is attached to a Workspace

8951 it cannot be detached or replaced. To rotate key material, rotate the underlying

8952 key on your cloud KMS; the `external_key_id` stays the same.

88798953

88808954- `tags: optional map[string] or null`

88818955

88828956 User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.

88838957

8958- `user_profile_id: optional string or null`

8959

8960 ID of a user profile to attach to the Workspace. Every inference request through the Workspace must then name this user profile; the API rejects a request that names none or a different one. The user profile can't be replaced or removed later; only deleting the user profile clears it.

8961

88848962#### Returns

88858963

88868964- `BetaWorkspace object`

from line 8983

89058983

89068984 - `compartment_id: string`

89078985

8908 Identifier for this Workspace's encryption compartment. When you configure a

8909 customer-managed encryption key (CMEK) on AWS, reference this value in your

8910 KMS key-policy condition so the key is scoped to this compartment. On GCP and

8911 Azure, Anthropic enforces the compartment binding automatically; you do not

8912 need to reference this value in your key configuration. See the CMEK

8913 integration guide for the required key configuration; unless your organization

8914 is on Claude Platform on AWS, it includes a separate value used during key

8915 validation. On Claude Platform on AWS there is no separate validation value:

8916 the key is validated against this Workspace's own value when it is attached, so

8917 if your key policy uses the compartment condition, add this value to it before

8918 attaching the key.

8986 Identifier for this Workspace's encryption compartment. When you configure an

8987 AWS KMS customer-managed encryption key (CMEK) that isn't bound to a user

8988 profile, reference this value in your KMS key-policy condition so the key is

8989 scoped to this compartment. On GCP and Azure, Anthropic enforces the compartment

8990 binding automatically; you do not need to reference this value in your key

8991 configuration. See the CMEK integration guide for the required key

8992 configuration; unless your organization is on Claude Platform on AWS, it

8993 includes a separate value used during key validation. On Claude Platform on AWS

8994 there is no separate validation value: your organization's own key is validated

8995 against this Workspace's own value when it is attached, so if its key policy

8996 uses the compartment condition, add this value to it before attaching the key.

89198997

89208998 - `created_at: string`

89218999

from line 9039

89619039 Workspace. Setting this field requires CMEK to be enabled for your

89629040 organization. When set, data stored for this Workspace is encrypted with the

89639041 referenced key. Create key configurations with the External Keys API. On

8964 Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a

8965 single-Region key in the same AWS account and Region as the Workspace. On that

8966 platform the key is validated against this Workspace when it is attached, so a

8967 key-policy problem is reported as an error on this request. This field is write-once:

8968 once a key is attached to a Workspace it cannot be detached or replaced. To

8969 rotate key material, rotate the underlying key on your cloud KMS; the

8970 `external_key_id` stays the same.

9042 Claude Platform on AWS the value for an AWS KMS key is the key's ARN, and your

9043 organization's own key must be a single-Region key in the same AWS account and

9044 Region as the Workspace. On that platform the key is validated against this

9045 Workspace when it is attached, so a key-policy problem is reported as an error

9046 on this request. This field is write-once: once a key is attached to a Workspace

9047 it cannot be detached or replaced. To rotate key material, rotate the underlying

9048 key on your cloud KMS; the `external_key_id` stays the same.

89719049

89729050 - `name: string`

89739051

from line 9055

89779055

89789056 User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.

89799057

9058 - `user_profile_id: string or null`

9059

9060 ID of the user profile attached to the Workspace, or null if it has none. Every inference request through a Workspace that has a user profile must name that profile.

9061

89809062#### Example

89819063

89829064```bash

from line 9073

89919073 "tags": {

89929074 "env": "prod",

89939075 "team": "platform"

8994 }

9076 },

9077 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

89959078 }'

89969079```

89979080

from line 9098

90159098 "env": "prod",

90169099 "team": "platform"

90179100 },

9018 "type": "workspace"

9101 "type": "workspace",

9102 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

90199103}

90209104```

90219105

from line 9147

90639147

90649148 - `compartment_id: string`

90659149

9066 Identifier for this Workspace's encryption compartment. When you configure a

9067 customer-managed encryption key (CMEK) on AWS, reference this value in your

9068 KMS key-policy condition so the key is scoped to this compartment. On GCP and

9069 Azure, Anthropic enforces the compartment binding automatically; you do not

9070 need to reference this value in your key configuration. See the CMEK

9071 integration guide for the required key configuration; unless your organization

9072 is on Claude Platform on AWS, it includes a separate value used during key

9073 validation. On Claude Platform on AWS there is no separate validation value:

9074 the key is validated against this Workspace's own value when it is attached, so

9075 if your key policy uses the compartment condition, add this value to it before

9076 attaching the key.

9150 Identifier for this Workspace's encryption compartment. When you configure an

9151 AWS KMS customer-managed encryption key (CMEK) that isn't bound to a user

9152 profile, reference this value in your KMS key-policy condition so the key is

9153 scoped to this compartment. On GCP and Azure, Anthropic enforces the compartment

9154 binding automatically; you do not need to reference this value in your key

9155 configuration. See the CMEK integration guide for the required key

9156 configuration; unless your organization is on Claude Platform on AWS, it

9157 includes a separate value used during key validation. On Claude Platform on AWS

9158 there is no separate validation value: your organization's own key is validated

9159 against this Workspace's own value when it is attached, so if its key policy

9160 uses the compartment condition, add this value to it before attaching the key.

90779161

90789162 - `created_at: string`

90799163

from line 9203

91199203 Workspace. Setting this field requires CMEK to be enabled for your

91209204 organization. When set, data stored for this Workspace is encrypted with the

91219205 referenced key. Create key configurations with the External Keys API. On

9122 Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a

9123 single-Region key in the same AWS account and Region as the Workspace. On that

9124 platform the key is validated against this Workspace when it is attached, so a

9125 key-policy problem is reported as an error on this request. This field is write-once:

9126 once a key is attached to a Workspace it cannot be detached or replaced. To

9127 rotate key material, rotate the underlying key on your cloud KMS; the

9128 `external_key_id` stays the same.

9206 Claude Platform on AWS the value for an AWS KMS key is the key's ARN, and your

9207 organization's own key must be a single-Region key in the same AWS account and

9208 Region as the Workspace. On that platform the key is validated against this

9209 Workspace when it is attached, so a key-policy problem is reported as an error

9210 on this request. This field is write-once: once a key is attached to a Workspace

9211 it cannot be detached or replaced. To rotate key material, rotate the underlying

9212 key on your cloud KMS; the `external_key_id` stays the same.

91299213

91309214 - `name: string`

91319215

from line 9219

91359219

91369220 User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.

91379221

9222 - `user_profile_id: string or null`

9223

9224 ID of the user profile attached to the Workspace, or null if it has none. Every inference request through a Workspace that has a user profile must name that profile.

9225

91389226#### Example

91399227

91409228```bash

from line 9251

91639251 "env": "prod",

91649252 "team": "platform"

91659253 },

9166 "type": "workspace"

9254 "type": "workspace",

9255 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

91679256}

91689257```

91699258

from line 9312

92239312 Workspace. Setting this field requires CMEK to be enabled for your

92249313 organization. When set, data stored for this Workspace is encrypted with the

92259314 referenced key. Create key configurations with the External Keys API. On

9226 Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a

9227 single-Region key in the same AWS account and Region as the Workspace. On that

9228 platform the key is validated against this Workspace when it is attached, so a

9229 key-policy problem is reported as an error on this request. This field is write-once:

9230 once a key is attached to a Workspace it cannot be detached or replaced. To

9231 rotate key material, rotate the underlying key on your cloud KMS; the

9232 `external_key_id` stays the same.

9315 Claude Platform on AWS the value for an AWS KMS key is the key's ARN, and your

9316 organization's own key must be a single-Region key in the same AWS account and

9317 Region as the Workspace. On that platform the key is validated against this

9318 Workspace when it is attached, so a key-policy problem is reported as an error

9319 on this request. This field is write-once: once a key is attached to a Workspace

9320 it cannot be detached or replaced. To rotate key material, rotate the underlying

9321 key on your cloud KMS; the `external_key_id` stays the same.

92339322

92349323- `name: optional string`

92359324

from line 9330

92419330

92429331 User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.

92439332

9333- `user_profile_id: optional string`

9334

9335 ID of a user profile to attach to the Workspace. Every inference request through the Workspace must then name this user profile; the API rejects a request that names none or a different one. Once set, it can't be replaced or removed (sending the same ID again changes nothing); only deleting the user profile clears it. The default Workspace can't have a user profile.

9336

92449337#### Returns

92459338

92469339- `BetaWorkspace object`

from line 9358

92659358

92669359 - `compartment_id: string`

92679360

9268 Identifier for this Workspace's encryption compartment. When you configure a

9269 customer-managed encryption key (CMEK) on AWS, reference this value in your

9270 KMS key-policy condition so the key is scoped to this compartment. On GCP and

9271 Azure, Anthropic enforces the compartment binding automatically; you do not

9272 need to reference this value in your key configuration. See the CMEK

9273 integration guide for the required key configuration; unless your organization

9274 is on Claude Platform on AWS, it includes a separate value used during key

9275 validation. On Claude Platform on AWS there is no separate validation value:

9276 the key is validated against this Workspace's own value when it is attached, so

9277 if your key policy uses the compartment condition, add this value to it before

9278 attaching the key.

9361 Identifier for this Workspace's encryption compartment. When you configure an

9362 AWS KMS customer-managed encryption key (CMEK) that isn't bound to a user

9363 profile, reference this value in your KMS key-policy condition so the key is

9364 scoped to this compartment. On GCP and Azure, Anthropic enforces the compartment

9365 binding automatically; you do not need to reference this value in your key

9366 configuration. See the CMEK integration guide for the required key

9367 configuration; unless your organization is on Claude Platform on AWS, it

9368 includes a separate value used during key validation. On Claude Platform on AWS

9369 there is no separate validation value: your organization's own key is validated

9370 against this Workspace's own value when it is attached, so if its key policy

9371 uses the compartment condition, add this value to it before attaching the key.

92799372

92809373 - `created_at: string`

92819374

from line 9414

93219414 Workspace. Setting this field requires CMEK to be enabled for your

93229415 organization. When set, data stored for this Workspace is encrypted with the

93239416 referenced key. Create key configurations with the External Keys API. On

9324 Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a

9325 single-Region key in the same AWS account and Region as the Workspace. On that

9326 platform the key is validated against this Workspace when it is attached, so a

9327 key-policy problem is reported as an error on this request. This field is write-once:

9328 once a key is attached to a Workspace it cannot be detached or replaced. To

9329 rotate key material, rotate the underlying key on your cloud KMS; the

9330 `external_key_id` stays the same.

9417 Claude Platform on AWS the value for an AWS KMS key is the key's ARN, and your

9418 organization's own key must be a single-Region key in the same AWS account and

9419 Region as the Workspace. On that platform the key is validated against this

9420 Workspace when it is attached, so a key-policy problem is reported as an error

9421 on this request. This field is write-once: once a key is attached to a Workspace

9422 it cannot be detached or replaced. To rotate key material, rotate the underlying

9423 key on your cloud KMS; the `external_key_id` stays the same.

93319424

93329425 - `name: string`

93339426

from line 9430

93379430

93389431 User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.

93399432

9433 - `user_profile_id: string or null`

9434

9435 ID of the user profile attached to the Workspace, or null if it has none. Every inference request through a Workspace that has a user profile must name that profile.

9436

93409437#### Example

93419438

93429439```bash

from line 9447

93509447 "tags": {

93519448 "env": "prod",

93529449 "team": "platform"

9353 }

9450 },

9451 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

93549452 }'

93559453```

93569454

from line 9472

93749472 "env": "prod",

93759473 "team": "platform"

93769474 },

9377 "type": "workspace"

9475 "type": "workspace",

9476 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

93789477}

93799478```

93809479

from line 9519

94209519

94219520 - `compartment_id: string`

94229521

9423 Identifier for this Workspace's encryption compartment. When you configure a

9424 customer-managed encryption key (CMEK) on AWS, reference this value in your

9425 KMS key-policy condition so the key is scoped to this compartment. On GCP and

9426 Azure, Anthropic enforces the compartment binding automatically; you do not

9427 need to reference this value in your key configuration. See the CMEK

9428 integration guide for the required key configuration; unless your organization

9429 is on Claude Platform on AWS, it includes a separate value used during key

9430 validation. On Claude Platform on AWS there is no separate validation value:

9431 the key is validated against this Workspace's own value when it is attached, so

9432 if your key policy uses the compartment condition, add this value to it before

9433 attaching the key.

9522 Identifier for this Workspace's encryption compartment. When you configure an

9523 AWS KMS customer-managed encryption key (CMEK) that isn't bound to a user

9524 profile, reference this value in your KMS key-policy condition so the key is

9525 scoped to this compartment. On GCP and Azure, Anthropic enforces the compartment

9526 binding automatically; you do not need to reference this value in your key

9527 configuration. See the CMEK integration guide for the required key

9528 configuration; unless your organization is on Claude Platform on AWS, it

9529 includes a separate value used during key validation. On Claude Platform on AWS

9530 there is no separate validation value: your organization's own key is validated

9531 against this Workspace's own value when it is attached, so if its key policy

9532 uses the compartment condition, add this value to it before attaching the key.

94349533

94359534 - `created_at: string`

94369535

from line 9575

94769575 Workspace. Setting this field requires CMEK to be enabled for your

94779576 organization. When set, data stored for this Workspace is encrypted with the

94789577 referenced key. Create key configurations with the External Keys API. On

9479 Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a

9480 single-Region key in the same AWS account and Region as the Workspace. On that

9481 platform the key is validated against this Workspace when it is attached, so a

9482 key-policy problem is reported as an error on this request. This field is write-once:

9483 once a key is attached to a Workspace it cannot be detached or replaced. To

9484 rotate key material, rotate the underlying key on your cloud KMS; the

9485 `external_key_id` stays the same.

9578 Claude Platform on AWS the value for an AWS KMS key is the key's ARN, and your

9579 organization's own key must be a single-Region key in the same AWS account and

9580 Region as the Workspace. On that platform the key is validated against this

9581 Workspace when it is attached, so a key-policy problem is reported as an error

9582 on this request. This field is write-once: once a key is attached to a Workspace

9583 it cannot be detached or replaced. To rotate key material, rotate the underlying

9584 key on your cloud KMS; the `external_key_id` stays the same.

94869585

94879586 - `name: string`

94889587

from line 9591

94929591

94939592 User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.

94949593

9594 - `user_profile_id: string or null`

9595

9596 ID of the user profile attached to the Workspace, or null if it has none. Every inference request through a Workspace that has a user profile must name that profile.

9597

94959598#### Example

94969599

94979600```bash

from line 9624

95219624 "env": "prod",

95229625 "team": "platform"

95239626 },

9524 "type": "workspace"

9627 "type": "workspace",

9628 "user_profile_id": "uprof_01SDCCSbTxrXDpWc1phhtcfK"

95259629}

95269630```

95279631

from line 10236

1013210236 "data": [

1013310237 {

1013410238 "type": "workspace_member",

10135 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

10239 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

1013610240 "workspace_id": "wrkspc_01JwQvzr7rXLA5AGx3HKfFUJ",

1013710241 "workspace_role": "workspace_admin"

1013810242 }

from line 10327

1022310327 -H 'anthropic-version: 2026-10-21' \

1022410328 -H "Authorization: Bearer $ANTHROPIC_API_KEY" \

1022510329 -d '{

10226 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

10330 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

1022710331 "workspace_role": "workspace_admin"

1022810332 }'

1022910333```

from line 10337

1023310337```json

1023410338{

1023510339 "type": "workspace_member",

10236 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

10340 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

1023710341 "workspace_id": "wrkspc_01JwQvzr7rXLA5AGx3HKfFUJ",

1023810342 "workspace_role": "workspace_admin"

1023910343}

from line 10414

1031010414```json

1031110415{

1031210416 "type": "workspace_member",

10313 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

10417 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

1031410418 "workspace_id": "wrkspc_01JwQvzr7rXLA5AGx3HKfFUJ",

1031510419 "workspace_role": "workspace_admin"

1031610420}

from line 10511

1040710511```json

1040810512{

1040910513 "type": "workspace_member",

10410 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

10514 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

1041110515 "workspace_id": "wrkspc_01JwQvzr7rXLA5AGx3HKfFUJ",

1041210516 "workspace_role": "workspace_admin"

1041310517}

from line 10573

1046910573```json

1047010574{

1047110575 "type": "workspace_member_deleted",

10472 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

10576 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

1047310577 "workspace_id": "wrkspc_01JwQvzr7rXLA5AGx3HKfFUJ"

1047410578}

1047510579```

from line 10735

1063110735

1063210736 - `"analytics-usage-cost-2026-08-24"`

1063310737

10738 - `"experimental-claude-tag-2026-10-21"`

10739

1063410740#### Returns

1063510741

1063610742- `data: array of BetaServiceAccountWorkspaceMember`

from line 10950

1084410950

1084510951 - `"analytics-usage-cost-2026-08-24"`

1084610952

10953 - `"experimental-claude-tag-2026-10-21"`

10954

1084710955#### Body parameters

1084810956

1084910957- `service_account_id: string`

from line 11181

1107311181

1107411182 - `"analytics-usage-cost-2026-08-24"`

1107511183

11184 - `"experimental-claude-tag-2026-10-21"`

11185

1107611186#### Returns

1107711187

1107811188- `BetaServiceAccountWorkspaceMember object`

from line 11389

1127911389

1128011390 - `"analytics-usage-cost-2026-08-24"`

1128111391

11392 - `"experimental-claude-tag-2026-10-21"`

11393

1128211394#### Body parameters

1128311395

1128411396- `workspace_role: BetaNoBillingWorkspaceRole`

from line 11614

1150211614

1150311615 - `"analytics-usage-cost-2026-08-24"`

1150411616

11617 - `"experimental-claude-tag-2026-10-21"`

11618

1150511619#### Returns

1150611620

1150711621- `type: "service_account_workspace_member_deleted"`

from line 11931

1181711931

1181811932Setting `state` to `enabled` turns on the Compliance API and begins

1181911933capturing organization activity events. Setting it to `disabled` turns

11820both off. `state` reflects whether the Compliance API is enabled.

11934both off. Access Transparency events are an exception: they are still

11935captured while the Compliance API is disabled, and the Compliance API

11936activity feed returns them once it is enabled again. `state` reflects

11937whether the Compliance API is enabled.

1182111938

1182211939A request that sets `state` to its current value succeeds and leaves the

1182311940resource unchanged. A `disabled` request stays in effect until a later

from line 12272

1215512272

1215612273 - `"analytics-usage-cost-2026-08-24"`

1215712274

12275 - `"experimental-claude-tag-2026-10-21"`

12276

1215812277#### Returns

1215912278

1216012279- `BetaMessagesUsageReport object`

from line 12422

1230312422 "ending_at": "2025-08-02T00:00:00Z",

1230412423 "results": [

1230512424 {

12306 "account_id": "user_01WCz1FkmYMm4gnmykNKUu3Q",

12425 "account_id": "user_01WxYzAbCdEfGhJkMnPqRsTu",

1230712426 "api_key_id": "apikey_01Rj2N8SVvo6BePZj99NhmiT",

1230812427 "cache_creation": {

1230912428 "ephemeral_1h_input_tokens": 0,

from line 12897

1277812897

1277912898 - `"analytics-usage-cost-2026-08-24"`

1278012899

12900 - `"experimental-claude-tag-2026-10-21"`

12901

1278112902#### Returns

1278212903

1278312904- `BetaCostReport object`

from line 13213

1309213213

1309313214 - `"analytics-usage-cost-2026-08-24"`

1309413215

13216 - `"experimental-claude-tag-2026-10-21"`

13217

1309513218#### Returns

1309613219

1309713220- `data: array of BetaOrganizationTunnel`

from line 13427

1330413427

1330513428 - `"analytics-usage-cost-2026-08-24"`

1330613429

13430 - `"experimental-claude-tag-2026-10-21"`

13431

1330713432#### Returns

1330813433

1330913434- `BetaOrganizationTunnel object`

from line 13637

1351213637

1351313638 - `"analytics-usage-cost-2026-08-24"`

1351413639

13640 - `"experimental-claude-tag-2026-10-21"`

13641

1351513642#### Returns

1351613643

1351713644- `BetaOrganizationTunnel object`

from line 13848

1372113848

1372213849 - `"analytics-usage-cost-2026-08-24"`

1372313850

13851 - `"experimental-claude-tag-2026-10-21"`

13852

1372413853#### Returns

1372513854

1372613855- `BetaOrganizationTunnelToken object`

from line 14031

1390214031

1390314032 - `"analytics-usage-cost-2026-08-24"`

1390414033

14034 - `"experimental-claude-tag-2026-10-21"`

14035

1390514036#### Body parameters

1390614037

1390714038- `reason: optional string or null`

from line 14225

1409414225

1409514226 - `"analytics-usage-cost-2026-08-24"`

1409614227

14228 - `"experimental-claude-tag-2026-10-21"`

14229

1409714230#### Body parameters

1409814231

1409914232- `ca_certificate_pem: string`

from line 14465

1433214465

1433314466 - `"analytics-usage-cost-2026-08-24"`

1433414467

14468 - `"experimental-claude-tag-2026-10-21"`

14469

1433514470#### Returns

1433614471

1433714472- `data: array of BetaOrganizationTunnelCertificate`

from line 14683

1454814683

1454914684 - `"analytics-usage-cost-2026-08-24"`

1455014685

14686 - `"experimental-claude-tag-2026-10-21"`

14687

1455114688#### Returns

1455214689

1455314690- `BetaOrganizationTunnelCertificate object`

from line 14896

1475914896

1476014897 - `"analytics-usage-cost-2026-08-24"`

1476114898

14899 - `"experimental-claude-tag-2026-10-21"`

14900

1476214901#### Returns

1476314902

1476414903- `BetaOrganizationTunnelCertificate object`

from line 17411

1727217411

1727317412 - `"analytics-usage-cost-2026-08-24"`

1727417413

17414 - `"experimental-claude-tag-2026-10-21"`

17415

1727517416#### Returns

1727617417

1727717418- `data: array of BetaAnalyticsUsageItem`

from line 17723

1758217723 "email_address": "[email protected]",

1758317724 "name": "Jane Doe",

1758417725 "type": "user_actor",

17585 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

17726 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

1758617727 },

1758717728 "actor_type": "user_actor",

1758817729 "agent_id": "agent_01VHd3EAGJ6PLkP9LM7DDu3f",

from line 17756

1761517756 "server_tool_use": {

1761617757 "web_search_requests": 10

1761717758 },

17618 "total_tokens": 5377000,

17759 "total_tokens": 5375500,

1761917760 "uncached_input_tokens": 1284500

1762017761 }

1762117762 }

1762217763 ],

17623 "data_refreshed_at": "2019-12-27T18:11:19.117Z",

17764 "data_refreshed_at": "2026-10-01T04:02:11Z",

1762417765 "has_more": true,

1762517766 "next_page": "next_page"

1762617767}

from line 18193

1805218193

1805318194 - `"analytics-usage-cost-2026-08-24"`

1805418195

18196 - `"experimental-claude-tag-2026-10-21"`

18197

1805518198#### Returns

1805618199

1805718200- `data: array of BetaAnalyticsCostItem`

from line 18511

1836818511 "email_address": "[email protected]",

1836918512 "name": "Jane Doe",

1837018513 "type": "user_actor",

18371 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

18514 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

1837218515 },

1837318516 "actor_type": "user_actor",

1837418517 "agent_id": "agent_01VHd3EAGJ6PLkP9LM7DDu3f",

from line 18537

1839418537 "starting_at": "2026-09-01T00:00:00Z"

1839518538 }

1839618539 ],

18397 "data_refreshed_at": "2019-12-27T18:11:19.117Z",

18540 "data_refreshed_at": "2026-10-01T04:02:11Z",

1839818541 "has_more": true,

1839918542 "next_page": "next_page"

1840018543}

from line 18918

1877518918

1877618919[Legacy] Get per-user token usage across a date range.

1877718920

18778Returns usage per user, ranked by the chosen token metric; `group_by[]` or `bucket_width` splits a user across several rows. Use this to see which users consume the most tokens. Only usage attributed to a user account is included; usage with no user attributed to it, such as requests attributed only to an API key, is left out. Available to organizations on a Claude Enterprise plan. Requires an API key with the `read:analytics` scope.

18921Returns usage per user, ranked by the chosen token metric; `group_by[]` or `bucket_width` splits a user across several rows. Use this to see which users consume the most tokens. Only usage attributed to a user account is included; usage with no user attributed to it, such as Code Review usage and Claude Tag usage in Slack channels, is left out. Available to organizations on a Claude Enterprise plan. Requires an API key with the `read:analytics` scope.

1877918922

1878018923This endpoint is one of the legacy Claude Enterprise usage and cost reports. For new integrations, use [`GET /v1/organizations/analytics/usage`](https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list?sdk_languages=http) in the Usage and Cost Analytics API instead. To move an existing integration, see [Migrate from the Claude Enterprise usage and cost reports](https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api?sdk_languages=http#migrate-from-enterprise-analytics-reports).

1878118924

from line 19709

1956619709

1956719710[Legacy] Get per-user cost in USD across a date range.

1956819711

19569Returns cost per user, ranked by spend; `group_by[]` or `bucket_width` splits a user across several rows. Use this to see which users account for the most cost. Only cost attributed to a user account is included; cost with no user attributed to it, such as requests attributed only to an API key, is left out. Available to organizations on a Claude Enterprise plan. Requires an API key with the `read:analytics` scope.

19712Returns cost per user, ranked by spend; `group_by[]` or `bucket_width` splits a user across several rows. Use this to see which users account for the most cost. Only cost attributed to a user account is included; cost with no user attributed to it, such as Code Review usage and Claude Tag usage in Slack channels, is left out. Available to organizations on a Claude Enterprise plan. Requires an API key with the `read:analytics` scope.

1957019713

1957119714This endpoint is one of the legacy Claude Enterprise usage and cost reports. For new integrations, use [`GET /v1/organizations/analytics/cost`](https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list?sdk_languages=http) in the Usage and Cost Analytics API instead. To move an existing integration, see [Migrate from the Claude Enterprise usage and cost reports](https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api?sdk_languages=http#migrate-from-enterprise-analytics-reports).

1957219715

from line 20254

2011120254

2011220255 - `"analytics-usage-cost-2026-08-24"`

2011320256

20257 - `"experimental-claude-tag-2026-10-21"`

20258

2011420259#### Body parameters

2011520260

2011620261- `amount: string or null`

from line 20445

2030020445 "amount": "50000",

2030120446 "scope": {

2030220447 "type": "user",

20303 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

20448 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2030420449 },

2030520450 "period": "monthly"

2030620451 }'

from line 20463

2031820463 "period": "daily",

2031920464 "scope": {

2032020465 "type": "user",

20321 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

20466 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2032220467 },

2032320468 "type": "spend_limit",

2032420469 "updated_at": "2019-12-27T18:11:19.117Z"

from line 20636

2049120636 "period": "daily",

2049220637 "scope": {

2049320638 "type": "user",

20494 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

20639 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2049520640 },

2049620641 "type": "spend_limit",

2049720642 "updated_at": "2019-12-27T18:11:19.117Z"

from line 20867

2072220867

2072320868 - `"analytics-usage-cost-2026-08-24"`

2072420869

20870 - `"experimental-claude-tag-2026-10-21"`

20871

2072520872#### Returns

2072620873

2072720874- `data: array of BetaSpendLimit`

from line 21018

2087121018 "period": "daily",

2087221019 "scope": {

2087321020 "type": "user",

20874 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21021 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2087521022 },

2087621023 "type": "spend_limit",

2087721024 "updated_at": "2019-12-27T18:11:19.117Z"

from line 21271

2112421271 "email_address": "email_address",

2112521272 "name": "name",

2112621273 "type": "user_actor",

21127 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21274 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2112821275 },

2112921276 "amount": "50000",

2113021277 "currency": "USD",

from line 21279

2113221279 "period_to_date_spend": "12050.5",

2113321280 "scope": {

2113421281 "type": "user",

21135 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21282 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2113621283 },

2113721284 "source": {

2113821285 "type": "user",

21139 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21286 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2114021287 },

2114121288 "spend_limit_id": "spend_limit_id"

2114221289 }

from line 21579

2143221579 "email_address": "email_address",

2143321580 "name": "name",

2143421581 "type": "user_actor",

21435 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21582 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2143621583 },

2143721584 "created_at": "2019-12-27T18:11:19.117Z",

2143821585 "period": "daily",

from line 21589

2144221589 "email_address": "email_address",

2144321590 "name": "name",

2144421591 "type": "user_actor",

21445 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21592 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2144621593 },

2144721594 "spend_summary": {

2144821595 "actor": {

from line 21597

2145021597 "email_address": "email_address",

2145121598 "name": "name",

2145221599 "type": "user_actor",

21453 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21600 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2145421601 },

2145521602 "amount": "50000",

2145621603 "currency": "USD",

from line 21605

2145821605 "period_to_date_spend": "12050.5",

2145921606 "scope": {

2146021607 "type": "user",

21461 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21608 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2146221609 },

2146321610 "source": {

2146421611 "type": "user",

21465 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21612 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2146621613 },

2146721614 "spend_limit_id": "spend_limit_id"

2146821615 },

from line 21884

2173721884 "email_address": "email_address",

2173821885 "name": "name",

2173921886 "type": "user_actor",

21740 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21887 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2174121888 },

2174221889 "created_at": "2019-12-27T18:11:19.117Z",

2174321890 "period": "daily",

from line 21894

2174721894 "email_address": "email_address",

2174821895 "name": "name",

2174921896 "type": "user_actor",

21750 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21897 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2175121898 },

2175221899 "spend_summary": {

2175321900 "actor": {

from line 21902

2175521902 "email_address": "email_address",

2175621903 "name": "name",

2175721904 "type": "user_actor",

21758 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21905 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2175921906 },

2176021907 "amount": "50000",

2176121908 "currency": "USD",

from line 21910

2176321910 "period_to_date_spend": "12050.5",

2176421911 "scope": {

2176521912 "type": "user",

21766 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21913 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2176721914 },

2176821915 "source": {

2176921916 "type": "user",

21770 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

21917 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2177121918 },

2177221919 "spend_limit_id": "spend_limit_id"

2177321920 },

from line 22277

2213022277 "email_address": "email_address",

2213122278 "name": "name",

2213222279 "type": "user_actor",

22133 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22280 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2213422281 },

2213522282 "created_at": "2019-12-27T18:11:19.117Z",

2213622283 "period": "daily",

from line 22287

2214022287 "email_address": "email_address",

2214122288 "name": "name",

2214222289 "type": "user_actor",

22143 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22290 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2214422291 },

2214522292 "spend_limit": {

2214622293 "id": "id",

from line 22298

2215122298 "period": "daily",

2215222299 "scope": {

2215322300 "type": "user",

22154 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22301 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2215522302 },

2215622303 "type": "spend_limit",

2215722304 "updated_at": "2019-12-27T18:11:19.117Z"

from line 22309

2216222309 "email_address": "email_address",

2216322310 "name": "name",

2216422311 "type": "user_actor",

22165 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22312 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2216622313 },

2216722314 "amount": "50000",

2216822315 "currency": "USD",

from line 22317

2217022317 "period_to_date_spend": "12050.5",

2217122318 "scope": {

2217222319 "type": "user",

22173 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22320 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2217422321 },

2217522322 "source": {

2217622323 "type": "user",

22177 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22324 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2217822325 },

2217922326 "spend_limit_id": "spend_limit_id"

2218022327 },

from line 22599

2245222599 "email_address": "email_address",

2245322600 "name": "name",

2245422601 "type": "user_actor",

22455 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22602 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2245622603 },

2245722604 "created_at": "2019-12-27T18:11:19.117Z",

2245822605 "period": "daily",

from line 22609

2246222609 "email_address": "email_address",

2246322610 "name": "name",

2246422611 "type": "user_actor",

22465 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22612 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2246622613 },

2246722614 "spend_summary": {

2246822615 "actor": {

from line 22617

2247022617 "email_address": "email_address",

2247122618 "name": "name",

2247222619 "type": "user_actor",

22473 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22620 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2247422621 },

2247522622 "amount": "50000",

2247622623 "currency": "USD",

from line 22625

2247822625 "period_to_date_spend": "12050.5",

2247922626 "scope": {

2248022627 "type": "user",

22481 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22628 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2248222629 },

2248322630 "source": {

2248422631 "type": "user",

22485 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

22632 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2248622633 },

2248722634 "spend_limit_id": "spend_limit_id"

2248822635 },

from line 23188

2304123188 "email": "[email protected]",

2304223189 "rbac_group_id": "rbac_group_012rppKaSVsmTo6NqRDXQXNF",

2304323190 "type": "rbac_group_member",

23044 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

23191 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2304523192 }

2304623193 ],

2304723194 "has_more": false,

from line 23262

2311523262 -H 'anthropic-version: 2026-10-21' \

2311623263 -H "Authorization: Bearer $ANTHROPIC_API_KEY" \

2311723264 -d '{

23118 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

23265 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2311923266 }'

2312023267```

2312123268

from line 23274

2312723274 "email": "[email protected]",

2312823275 "rbac_group_id": "rbac_group_012rppKaSVsmTo6NqRDXQXNF",

2312923276 "type": "rbac_group_member",

23130 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

23277 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2313123278}

2313223279```

2313323280

from line 23335

2318823335{

2318923336 "rbac_group_id": "rbac_group_012rppKaSVsmTo6NqRDXQXNF",

2319023337 "type": "rbac_group_member_deleted",

23191 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

23338 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2319223339}

2319323340```

2319423341

from line 23569

2342223569 The vocabulary follows the resource: an `organization` grant carries a

2342323570 product-feature entitlement (for example `chat`), an admin-panel

2342423571 permission entitlement (`permission_*`), or a blanket capability-access

23425 mode — `capability_access_all` grants every product-feature entitlement,

23426 and `capability_access_all_ga` grants the generally-available subset as

23427 it stands at permission-check time; neither mode grants model-access

23428 entitlements. A consumer enumerating a role's per-feature grants should

23572 mode — `capability_access_all` grants every product-feature entitlement

23573 except `inference_hooks_inspection`, which a role holds only through its

23574 own row, and `capability_access_all_ga` grants the generally-available

23575 subset of those as it stands at permission-check time; neither mode grants

23576 model-access entitlements. A consumer enumerating a role's per-feature grants should

2342923577 treat a blanket row as granting every product-feature entitlement it

2343023578 covers, or it will under-report the role's effective access. A `connector_tool` grant carries

2343123579 a tool-access action (`use` or `always_allow`); a `connector_scope` grant

from line 23854

2370623854

2370723855 - `"analytics-usage-cost-2026-08-24"`

2370823856

23857 - `"experimental-claude-tag-2026-10-21"`

23858

2370923859#### Body parameters (form-data)

2371023860

2371123861- `files: array of file`

from line 24106

2395624106 "created_by": {

2395724107 "email_address": "[email protected]",

2395824108 "type": "user_actor",

23959 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

24109 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2396024110 },

2396124111 "description": "Reviews pull requests against your team's conventions.",

2396224112 "display_name": "Code Review Helper",

from line 24265

2411524265

2411624266 - `"analytics-usage-cost-2026-08-24"`

2411724267

24268 - `"experimental-claude-tag-2026-10-21"`

24269

2411824270#### Returns

2411924271

2412024272- `BetaPlugin object`

from line 24499

2434724499 "created_by": {

2434824500 "email_address": "[email protected]",

2434924501 "type": "user_actor",

24350 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

24502 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2435124503 },

2435224504 "description": "Reviews pull requests against your team's conventions.",

2435324505 "display_name": "Code Review Helper",

from line 24675

2452324675

2452424676 - `"analytics-usage-cost-2026-08-24"`

2452524677

24678 - `"experimental-claude-tag-2026-10-21"`

24679

2452624680#### Body parameters

2452724681

2452824682- `served_version_id: string`

from line 24919

2476524919 "created_by": {

2476624920 "email_address": "[email protected]",

2476724921 "type": "user_actor",

24768 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

24922 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2476924923 },

2477024924 "description": "Reviews pull requests against your team's conventions.",

2477124925 "display_name": "Code Review Helper",

from line 25132

2497825132

2497925133 - `"analytics-usage-cost-2026-08-24"`

2498025134

25135 - `"experimental-claude-tag-2026-10-21"`

25136

2498125137#### Returns

2498225138

2498325139- `data: array of BetaPlugin`

from line 25372

2521625372 "created_by": {

2521725373 "email_address": "[email protected]",

2521825374 "type": "user_actor",

25219 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

25375 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2522025376 },

2522125377 "description": "Reviews pull requests against your team's conventions.",

2522225378 "display_name": "Code Review Helper",

from line 25539

2538325539

2538425540 - `"analytics-usage-cost-2026-08-24"`

2538525541

25542 - `"experimental-claude-tag-2026-10-21"`

25543

2538625544#### Returns

2538725545

2538825546- `BetaDeletedPlugin object`

from line 25727

2556925727

2557025728 - `"analytics-usage-cost-2026-08-24"`

2557125729

25730 - `"experimental-claude-tag-2026-10-21"`

25731

2557225732#### Body parameters (form-data)

2557325733

2557425734- `files: array of file`

from line 25913

2575325913 "created_by": {

2575425914 "email_address": "[email protected]",

2575525915 "type": "user_actor",

25756 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

25916 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2575725917 },

2575825918 "description": "Reviews pull requests against your team's conventions.",

2575925919 "display_name": "Code Review Helper",

from line 26080

2592026080

2592126081 - `"analytics-usage-cost-2026-08-24"`

2592226082

26083 - `"experimental-claude-tag-2026-10-21"`

26084

2592326085#### Returns

2592426086

2592526087- `data: array of BetaPluginVersion`

from line 26258

2609626258 "created_by": {

2609726259 "email_address": "[email protected]",

2609826260 "type": "user_actor",

26099 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

26261 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2610026262 },

2610126263 "description": "Reviews pull requests against your team's conventions.",

2610226264 "display_name": "Code Review Helper",

from line 26415

2625326415

2625426416 - `"analytics-usage-cost-2026-08-24"`

2625526417

26418 - `"experimental-claude-tag-2026-10-21"`

26419

2625626420#### Returns

2625726421

2625826422- `BetaPluginVersion object`

from line 26587

2642326587 "created_by": {

2642426588 "email_address": "[email protected]",

2642526589 "type": "user_actor",

26426 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

26590 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

2642726591 },

2642826592 "description": "Reviews pull requests against your team's conventions.",

2642926593 "display_name": "Code Review Helper",

from line 26757

2659326757

2659426758 - `"analytics-usage-cost-2026-08-24"`

2659526759

26760 - `"experimental-claude-tag-2026-10-21"`

26761

2659626762#### Example

2659726763

2659826764```bash

from line 26935

2676926935

2677026936 - `"analytics-usage-cost-2026-08-24"`

2677126937

26938 - `"experimental-claude-tag-2026-10-21"`

26939

2677226940#### Returns

2677326941

2677426942- `data: array of BetaPluginInstallationSetting`

from line 27192

2702427192

2702527193 - `"analytics-usage-cost-2026-08-24"`

2702627194

27195 - `"experimental-claude-tag-2026-10-21"`

27196

2702727197#### Body parameters

2702827198

2702927199- `installation_preference: "auto_install" or "available" or "not_available" or "required"`

from line 27466

2729627466

2729727467 - `"analytics-usage-cost-2026-08-24"`

2729827468

27469 - `"experimental-claude-tag-2026-10-21"`

27470

2729927471#### Returns

2730027472

2730127473- `BetaDeletedPluginInstallationSetting object`

from line 27714

2754227714

2754327715 - `"analytics-usage-cost-2026-08-24"`

2754427716

27717 - `"experimental-claude-tag-2026-10-21"`

27718

2754527719#### Returns

2754627720

2754727721- `data: array of BetaPluginShare`

from line 27979

2780527979

2780627980 - `"analytics-usage-cost-2026-08-24"`

2780727981

27982 - `"experimental-claude-tag-2026-10-21"`

27983

2780827984#### Returns

2780927985

2781027986- `data: array of BetaPluginMarketplace`

from line 28256

2808028256

2808128257 - `"analytics-usage-cost-2026-08-24"`

2808228258

28259 - `"experimental-claude-tag-2026-10-21"`

28260

2808328261#### Returns

2808428262

2808528263- `BetaPluginMarketplace object`

from line 28523

2834528523

2834628524 - `"analytics-usage-cost-2026-08-24"`

2834728525

28526 - `"experimental-claude-tag-2026-10-21"`

28527

2834828528#### Body parameters

2834928529

2835028530- `default_installation_preference: "auto_install" or "available" or "not_available" or "required"`

from line 28812

2863228812

2863328813 - `"analytics-usage-cost-2026-08-24"`

2863428814

28815 - `"experimental-claude-tag-2026-10-21"`

28816

2863528817#### Body parameters

2863628818

2863728819- `repository_url: string`

from line 29088

2890629088

2890729089 - `"analytics-usage-cost-2026-08-24"`

2890829090

29091 - `"experimental-claude-tag-2026-10-21"`

29092

2890929093#### Body parameters (form-data)

2891029094

2891129095- `archive: file`

from line 29348

2916429348

2916529349 - `"analytics-usage-cost-2026-08-24"`

2916629350

29351 - `"experimental-claude-tag-2026-10-21"`

29352

2916729353#### Body parameters

2916829354

2916929355- `endpoint_url: string`

from line 29829

2964329829

2964429830 - `"analytics-usage-cost-2026-08-24"`

2964529831

29832 - `"experimental-claude-tag-2026-10-21"`

29833

2964629834#### Returns

2964729835

2964829836- `data: array of BetaTelemetryDestination`

from line 30188

3000030188

3000130189 - `"analytics-usage-cost-2026-08-24"`

3000230190

30191 - `"experimental-claude-tag-2026-10-21"`

30192

3000330193#### Returns

3000430194

3000530195- `BetaTelemetryDestination object`

from line 30542

3035230542

3035330543 - `"analytics-usage-cost-2026-08-24"`

3035430544

30545 - `"experimental-claude-tag-2026-10-21"`

30546

3035530547#### Body parameters

3035630548

3035730549- `display_name: optional string or null`

from line 31030

3083831030

3083931031 - `"analytics-usage-cost-2026-08-24"`

3084031032

31033 - `"experimental-claude-tag-2026-10-21"`

31034

3084131035#### Returns

3084231036

3084331037- `BetaDeletedTelemetryDestination object`

from line 31197

3100331197

3100431198 - `"analytics-usage-cost-2026-08-24"`

3100531199

31200 - `"experimental-claude-tag-2026-10-21"`

31201

3100631202#### Returns

3100731203

3100831204- `BetaTelemetryDestinationTestResponse object`

from line 31400

3120431400

3120531401 - `"analytics-usage-cost-2026-08-24"`

3120631402

31403 - `"experimental-claude-tag-2026-10-21"`

31404

3120731405#### Returns

3120831406

3120931407- `data: array of BetaTelemetryDestinationWorkspace`

from line 31597

3139931597

3140031598 - `"analytics-usage-cost-2026-08-24"`

3140131599

31600 - `"experimental-claude-tag-2026-10-21"`

31601

3140231602#### Returns

3140331603

3140431604- `BetaTelemetryDestinationWorkspace object`

from line 31788

3158831788

3158931789 - `"analytics-usage-cost-2026-08-24"`

3159031790

31791 - `"experimental-claude-tag-2026-10-21"`

31792

3159131793#### Returns

3159231794

3159331795- `BetaDeletedTelemetryDestinationWorkspace object`

from line 31964

3176231964

3176331965 - `"analytics-usage-cost-2026-08-24"`

3176431966

31967 - `"experimental-claude-tag-2026-10-21"`

31968

3176531969#### Body parameters

3176631970

3176731971- `enabled_events: array of "user_profile.enforcement_action_opened" or "user_profile.enforcement_action_resolved"`

from line 32240

3203632240

3203732241 - `"analytics-usage-cost-2026-08-24"`

3203832242

32243 - `"experimental-claude-tag-2026-10-21"`

32244

3203932245#### Returns

3204032246

3204132247- `data: array of BetaOrganizationWebhookEndpoint`

from line 32480

3227432480

3227532481 - `"analytics-usage-cost-2026-08-24"`

3227632482

32483 - `"experimental-claude-tag-2026-10-21"`

32484

3227732485#### Returns

3227832486

3227932487- `BetaOrganizationWebhookEndpoint object`

from line 32702

3249432702

3249532703 - `"analytics-usage-cost-2026-08-24"`

3249632704

32705 - `"experimental-claude-tag-2026-10-21"`

32706

3249732707#### Body parameters

3249832708

3249932709- `description: optional string`

from line 32958

3274832958

3274932959 - `"analytics-usage-cost-2026-08-24"`

3275032960

32961 - `"experimental-claude-tag-2026-10-21"`

32962

3275132963#### Returns

3275232964

3275332965- `BetaDeletedOrganizationWebhookEndpoint object`

from line 33134

3292233134

3292333135 - `"analytics-usage-cost-2026-08-24"`

3292433136

33137 - `"experimental-claude-tag-2026-10-21"`

33138

3292533139#### Returns

3292633140

3292733141- `BetaOrganizationWebhookEndpointSigningSecret object`

from line 33163

3294933163 "signing_secret": "signing_secret"

3295033164}

3295133165```

33166

33167## Organization › Claude Tag › Slack Channels

33168

33169### Set Up Slack Channel

33170

33171**POST** `/v1/organizations/claude_tag/slack_channels`

33172

33173Set a Slack channel up for Claude Tag and return it with its identifier. If the channel is already set up, whether in claude.ai, by Claude's first use in it or by an earlier request, the existing channel is returned, so the request can be repeated. `settings`, when sent, is applied as an update applies it. Before a channel is set up, Claude's Slack app confirms with Slack that the channel is in the workspace, is not archived and is visible to the app; a channel that fails the confirmation returns 404. Setting the channel up and applying `settings` are two steps: when the second is refused, the response is the refusal, the channel stays set up with every setting unset, and the error's `details.slack_channel_id` holds the channel's identifier.

33174

33175#### Headers

33176

33177- `"anthropic-version": optional string`

33178

33179 The version of the Claude API you want to use.

33180

33181 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

33182

33183- `"anthropic-beta": optional array of AnthropicBeta`

33184

33185 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

33186

33187 - `string`

33188

33189 - `"message-batches-2024-09-24"`

33190

33191 - `"prompt-caching-2024-07-31"`

33192

33193 - `"computer-use-2024-10-22"`

33194

33195 - `"computer-use-2025-01-24"`

33196

33197 - `"pdfs-2024-09-25"`

33198

33199 - `"token-counting-2024-11-01"`

33200

33201 - `"token-efficient-tools-2025-02-19"`

33202

33203 - `"output-128k-2025-02-19"`

33204

33205 - `"files-api-2025-04-14"`

33206

33207 - `"mcp-client-2025-04-04"`

33208

33209 - `"mcp-client-2025-11-20"`

33210

33211 - `"dev-full-thinking-2025-05-14"`

33212

33213 - `"interleaved-thinking-2025-05-14"`

33214

33215 - `"code-execution-2025-05-22"`

33216

33217 - `"extended-cache-ttl-2025-04-11"`

33218

33219 - `"context-1m-2025-08-07"`

33220

33221 - `"context-management-2025-06-27"`

33222

33223 - `"model-context-window-exceeded-2025-08-26"`

33224

33225 - `"skills-2025-10-02"`

33226

33227 - `"fast-mode-2026-02-01"`

33228

33229 - `"output-300k-2026-03-24"`

33230

33231 - `"user-profiles-2026-03-24"`

33232

33233 - `"user-profiles-2026-08-18"`

33234

33235 - `"user-profiles-2026-09-04"`

33236

33237 - `"advisor-tool-2026-03-01"`

33238

33239 - `"managed-agents-2026-04-01"`

33240

33241 - `"cache-diagnosis-2026-04-07"`

33242

33243 - `"dreaming-2026-04-21"`

33244

33245 - `"thinking-token-count-2026-05-13"`

33246

33247 - `"server-side-fallback-2026-06-01"`

33248

33249 - `"server-side-fallback-2026-07-01"`

33250

33251 - `"fallback-credit-2026-06-01"`

33252

33253 - `"fallback-credit-2026-07-01"`

33254

33255 - `"agent-memory-2026-07-22"`

33256

33257 - `"mid-conversation-tool-changes-2026-07-01"`

33258

33259 - `"compact-2026-01-12"`

33260

33261 - `"computer-use-2025-11-24"`

33262

33263 - `"mcp-tunnels-2026-06-22"`

33264

33265 - `"structured-outputs-2025-11-13"`

33266

33267 - `"task-budgets-2026-03-13"`

33268

33269 - `"thinking-display-updates-2026-08-18"`

33270

33271 - `"ce-user-management-2026-07-13"`

33272

33273 - `"mid-conversation-output-config-2026-07-01"`

33274

33275 - `"thinking-binding-controls-2026-08-01"`

33276

33277 - `"mid-conversation-system-clear-at-2026-08-21"`

33278

33279 - `"compact-2026-09-04"`

33280

33281 - `"inline-tools-2026-09-15"`

33282

33283 - `"mcp-client-2026-09-15"`

33284

33285 - `"ce-plugins-2026-09-01"`

33286

33287 - `"spend-limit-reads-2026-09-26"`

33288

33289 - `"telemetry-destinations-2026-08-11"`

33290

33291 - `"webhooks-2026-10-07"`

33292

33293 - `"message-threads-2026-08-12"`

33294

33295 - `"analytics-usage-cost-2026-08-24"`

33296

33297 - `"experimental-claude-tag-2026-10-21"`

33298

33299#### Body parameters

33300

33301- `slack: BetaClaudeTagSlackChannelIdentifiers`

33302

33303 Slack's own identifiers for the channel to set up. The organization must own the workspace.

33304

33305 - `channel_id: string`

33306

33307 Slack's identifier for the channel, starting with `C` or `G`, such as `C0456EFGH`.

33308

33309 minLength: 2, maxLength: 32

33310

33311 - `workspace_id: string`

33312

33313 Slack's identifier for the workspace the channel is in, starting with `T`, such as `T0123ABCD`.

33314

33315 minLength: 2, maxLength: 32

33316

33317- `settings: optional BetaClaudeTagSettingsCreate`

33318

33319 Settings to store on the channel, applied as Update Slack Channel applies them.

33320

33321 - `enabled: optional boolean or null`

33322

33323 Whether Claude is on. `true` on a channel turns Claude on in it even when its workspace stores `false`. Null clears it.

33324

33325 - `environment_id: optional string or null`

33326

33327 The environment Claude's sessions run in: the identifier of an active environment of the organization, starting with `env_`. Null clears it.

33328

33329 minLength: 1, maxLength: 64

33330

33331 - `guest_policy: optional BetaClaudeTagGuestPolicy or null`

33332

33333 What Claude does in a channel while a Slack guest is in it. Null clears it.

33334

33335 - `BetaClaudeTagGuestPolicyAllow object`

33336

33337 Claude answers in the channel.

33338

33339 - `type: "allow"`

33340

33341 - `BetaClaudeTagGuestPolicyRestrict object`

33342

33343 Claude does not answer in the channel.

33344

33345 - `type: "restrict"`

33346

33347 - `BetaClaudeTagGuestPolicyChannelOnly object`

33348

33349 Claude answers with only the Bundles attached to the channel itself.

33350

33351 - `type: "channel_only"`

33352

33353 - `instructions: optional string or null`

33354

33355 Instructions for Claude, up to 8,192 characters after leading and trailing white space is trimmed. Null or blank text clears them.

33356

33357 - `model: optional string or null`

33358

33359 The model Claude uses: the identifier of a model, or the name of a model line, that the organization is entitled to for Claude Tag. Null clears it.

33360

33361 minLength: 1, maxLength: 128

33362

33363#### Returns

33364

33365- `BetaClaudeTagSlackChannel object`

33366

33367 A Slack channel set up for Claude Tag, with the settings it stores or, with `include_inherited`, the settings that apply in it.

33368

33369 - `type: "slack_channel"`

33370

33371 - `id: string`

33372

33373 The channel's identifier. Starts with `slch_`. It does not change for as long as the channel is set up, with one exception: a channel that is not configured gets a new one when claude.ai replaces its record, which happens only while the organization's read-only mode is off. A channel that is removed and set up again also has a new one. The old one keeps returning 404; look the channel up by its Slack identifiers for the new one.

33374

33375 - `created_at: string`

33376

33377 Time the channel was set up, by whichever route. It does not change for as long as `id` does not.

33378

33379 format: date-time

33380

33381 - `settings: BetaClaudeTagSettings`

33382

33383 What the channel itself stores or, with `include_inherited`, what applies in it. `sources` says which.

33384

33385 - `enabled: boolean or null`

33386

33387 Whether Claude is on. The nearest settings level that sets it decides.

33388

33389 - `environment_id: string or null`

33390

33391 The environment Claude's sessions run in, starting with `env_`, or the identifier of a self-hosted runner pool, starting with `ccpool_`. The nearest settings level that sets it decides.

33392

33393 - `guest_policy: BetaClaudeTagGuestPolicy or null`

33394

33395 What Claude does in a channel while a Slack guest is in it. The nearest settings level that sets it decides.

33396

33397 - `BetaClaudeTagGuestPolicyAllow object`

33398

33399 Claude answers in the channel.

33400

33401 - `type: "allow"`

33402

33403 - `BetaClaudeTagGuestPolicyRestrict object`

33404

33405 Claude does not answer in the channel.

33406

33407 - `type: "restrict"`

33408

33409 - `BetaClaudeTagGuestPolicyChannelOnly object`

33410

33411 Claude answers with only the Bundles attached to the channel itself.

33412

33413 - `type: "channel_only"`

33414

33415 - `instructions: string or null`

33416

33417 Instructions for Claude. Instructions add up: the text of every settings level that sets it is joined, broadest first.

33418

33419 - `model: string or null`

33420

33421 The model Claude uses: a model identifier or the name of a model line. The nearest settings level that sets it decides. When none does, Claude uses the product's default model.

33422

33423 - `slack: BetaClaudeTagSlackChannelIdentifiers`

33424

33425 Slack's own identifiers for the channel.

33426

33427 - `channel_id: string`

33428

33429 Slack's identifier for the channel, starting with `C` or `G`, such as `C0456EFGH`.

33430

33431 minLength: 2, maxLength: 32

33432

33433 - `workspace_id: string`

33434

33435 Slack's identifier for the workspace the channel is in, starting with `T`, such as `T0123ABCD`.

33436

33437 minLength: 2, maxLength: 32

33438

33439 - `sources: BetaClaudeTagSettingSources`

33440

33441 For each setting, the settings level its value in `settings` comes from.

33442

33443 - `enabled: BetaClaudeTagSettingSource or null`

33444

33445 The settings level `enabled` comes from.

33446

33447 - `BetaClaudeTagSlackChannelSettingSource object`

33448

33449 The value comes from the addressed Slack channel itself.

33450

33451 - `type: "slack_channel"`

33452

33453 - `slack_channel_id: string`

33454

33455 The channel's identifier, starting with `slch_`.

33456

33457 - `BetaClaudeTagSlackWorkspaceSettingSource object`

33458

33459 The value comes from a Slack workspace of the organization.

33460

33461 - `type: "slack_workspace"`

33462

33463 - `slack_workspace_id: string`

33464

33465 The workspace's identifier, starting with `slws_`.

33466

33467 - `BetaClaudeTagSlackParentChannelSettingSource object`

33468

33469 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

33470

33471 - `type: "slack_parent_channel"`

33472

33473 - `slack_channel_id: string`

33474

33475 The parent channel's identifier, starting with `slch_`.

33476

33477 - `BetaClaudeTagSlackDefaultSettingSource object`

33478

33479 The value comes from the organization's Slack defaults, which have no identifier.

33480

33481 - `type: "slack_default"`

33482

33483 - `BetaClaudeTagDefaultSettingSource object`

33484

33485 No settings level sets a value, and the built-in default applies.

33486

33487 - `type: "default"`

33488

33489 - `BetaClaudeTagSlackGridChannelSettingSource object`

33490

33491 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

33492

33493 - `type: "slack_grid_channel"`

33494

33495 - `slack: BetaClaudeTagSlackChannelIdentifiers`

33496

33497 Slack's own identifiers for the channel under that other workspace.

33498

33499 - `environment_id: BetaClaudeTagSettingSource or null`

33500

33501 The settings level `environment_id` comes from.

33502

33503 - `guest_policy: BetaClaudeTagSettingSource or null`

33504

33505 The settings level `guest_policy` comes from.

33506

33507 - `instructions: array of BetaClaudeTagSettingSource`

33508

33509 Every settings level that contributes text to `instructions`, broadest first, in the order their text is joined. Empty when none does.

33510

33511 - `BetaClaudeTagSlackChannelSettingSource object`

33512

33513 The value comes from the addressed Slack channel itself.

33514

33515 - `BetaClaudeTagSlackWorkspaceSettingSource object`

33516

33517 The value comes from a Slack workspace of the organization.

33518

33519 - `BetaClaudeTagSlackParentChannelSettingSource object`

33520

33521 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

33522

33523 - `BetaClaudeTagSlackDefaultSettingSource object`

33524

33525 The value comes from the organization's Slack defaults, which have no identifier.

33526

33527 - `BetaClaudeTagDefaultSettingSource object`

33528

33529 No settings level sets a value, and the built-in default applies.

33530

33531 - `BetaClaudeTagSlackGridChannelSettingSource object`

33532

33533 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

33534

33535 - `model: BetaClaudeTagSettingSource or null`

33536

33537 The settings level `model` comes from.

33538

33539 - `updated_at: string`

33540

33541 Time Claude Tag's record of the channel last changed. It moves on every change to a setting, and can also move when something no response holds changes, so a later value says to read the channel again, not that a setting changed. Attaching or detaching a Bundle does not by itself move it.

33542

33543 format: date-time

33544

33545#### Example

33546

33547```bash

33548curl https://api.anthropic.com/v1/organizations/claude_tag/slack_channels \

33549 -H 'Content-Type: application/json' \

33550 -H 'anthropic-version: 2026-10-21' \

33551 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

33552 -H "Authorization: Bearer $ANTHROPIC_API_KEY" \

33553 -d '{

33554 "slack": {

33555 "channel_id": "xx",

33556 "workspace_id": "xx"

33557 }

33558 }'

33559```

33560

33561##### Response (200)

33562

33563```json

33564{

33565 "id": "id",

33566 "created_at": "2019-12-27T18:11:19.117Z",

33567 "settings": {

33568 "enabled": true,

33569 "environment_id": "environment_id",

33570 "guest_policy": {

33571 "type": "allow"

33572 },

33573 "instructions": "instructions",

33574 "model": "model"

33575 },

33576 "slack": {

33577 "channel_id": "xx",

33578 "workspace_id": "xx"

33579 },

33580 "sources": {

33581 "enabled": {

33582 "slack_channel_id": "slack_channel_id",

33583 "type": "slack_channel"

33584 },

33585 "environment_id": {

33586 "slack_channel_id": "slack_channel_id",

33587 "type": "slack_channel"

33588 },

33589 "guest_policy": {

33590 "slack_channel_id": "slack_channel_id",

33591 "type": "slack_channel"

33592 },

33593 "instructions": [

33594 {

33595 "slack_channel_id": "slack_channel_id",

33596 "type": "slack_channel"

33597 }

33598 ],

33599 "model": {

33600 "slack_channel_id": "slack_channel_id",

33601 "type": "slack_channel"

33602 }

33603 },

33604 "type": "slack_channel",

33605 "updated_at": "2019-12-27T18:11:19.117Z"

33606}

33607```

33608

33609### List Slack Channels

33610

33611**GET** `/v1/organizations/claude_tag/slack_channels`

33612

33613List the set-up Slack channels of one Slack workspace, newest first by `created_at`. A channel that is not set up is not listed: Slack holds the full list of a workspace's channels. Pass `slack_channel_id` as well to look one channel up by its Slack identifiers: `data` then holds that channel, or is empty when it is not set up, and `limit` and `page` are ignored. A workspace that the organization owns only through a Slack Enterprise Grid returns 404 until it is set up. With `include_inherited=true` each channel's `settings` holds what applies in it, and the request fails with 503 rather than resolve against a settings level that could not be read. A `page` cursor resumes only a request with the same `include_inherited`.

33614

33615#### Query parameters

33616

33617- `slack_workspace_id: string`

33618

33619 The Slack workspace whose channels to list: Slack's identifier for it, as in the channel object's `slack.workspace_id`, not the workspace object's `id`.

33620

33621- `include_inherited: optional boolean`

33622

33623 When `true`, `settings` holds what applies in each channel once the settings levels above it are taken into account, and every member of `sources` names the level a value comes from. When `false` or left out, `settings` holds what each channel itself stores.

33624

33625- `limit: optional number`

33626

33627 Number of items to return per page, between 1 and 100. Defaults to 20.

33628

33629 format: int32

33630

33631- `page: optional string`

33632

33633 Opaque cursor from a previous response's `next_page`.

33634

33635- `slack_channel_id: optional string`

33636

33637 Slack's identifier for one channel of that workspace, as in the channel object's `slack.channel_id`. It is not the channel object's `id`. With it `data` holds that channel alone, or nothing when the channel is not set up.

33638

33639#### Headers

33640

33641- `"anthropic-version": optional string`

33642

33643 The version of the Claude API you want to use.

33644

33645 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

33646

33647- `"anthropic-beta": optional array of AnthropicBeta`

33648

33649 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

33650

33651 - `string`

33652

33653 - `"message-batches-2024-09-24"`

33654

33655 - `"prompt-caching-2024-07-31"`

33656

33657 - `"computer-use-2024-10-22"`

33658

33659 - `"computer-use-2025-01-24"`

33660

33661 - `"pdfs-2024-09-25"`

33662

33663 - `"token-counting-2024-11-01"`

33664

33665 - `"token-efficient-tools-2025-02-19"`

33666

33667 - `"output-128k-2025-02-19"`

33668

33669 - `"files-api-2025-04-14"`

33670

33671 - `"mcp-client-2025-04-04"`

33672

33673 - `"mcp-client-2025-11-20"`

33674

33675 - `"dev-full-thinking-2025-05-14"`

33676

33677 - `"interleaved-thinking-2025-05-14"`

33678

33679 - `"code-execution-2025-05-22"`

33680

33681 - `"extended-cache-ttl-2025-04-11"`

33682

33683 - `"context-1m-2025-08-07"`

33684

33685 - `"context-management-2025-06-27"`

33686

33687 - `"model-context-window-exceeded-2025-08-26"`

33688

33689 - `"skills-2025-10-02"`

33690

33691 - `"fast-mode-2026-02-01"`

33692

33693 - `"output-300k-2026-03-24"`

33694

33695 - `"user-profiles-2026-03-24"`

33696

33697 - `"user-profiles-2026-08-18"`

33698

33699 - `"user-profiles-2026-09-04"`

33700

33701 - `"advisor-tool-2026-03-01"`

33702

33703 - `"managed-agents-2026-04-01"`

33704

33705 - `"cache-diagnosis-2026-04-07"`

33706

33707 - `"dreaming-2026-04-21"`

33708

33709 - `"thinking-token-count-2026-05-13"`

33710

33711 - `"server-side-fallback-2026-06-01"`

33712

33713 - `"server-side-fallback-2026-07-01"`

33714

33715 - `"fallback-credit-2026-06-01"`

33716

33717 - `"fallback-credit-2026-07-01"`

33718

33719 - `"agent-memory-2026-07-22"`

33720

33721 - `"mid-conversation-tool-changes-2026-07-01"`

33722

33723 - `"compact-2026-01-12"`

33724

33725 - `"computer-use-2025-11-24"`

33726

33727 - `"mcp-tunnels-2026-06-22"`

33728

33729 - `"structured-outputs-2025-11-13"`

33730

33731 - `"task-budgets-2026-03-13"`

33732

33733 - `"thinking-display-updates-2026-08-18"`

33734

33735 - `"ce-user-management-2026-07-13"`

33736

33737 - `"mid-conversation-output-config-2026-07-01"`

33738

33739 - `"thinking-binding-controls-2026-08-01"`

33740

33741 - `"mid-conversation-system-clear-at-2026-08-21"`

33742

33743 - `"compact-2026-09-04"`

33744

33745 - `"inline-tools-2026-09-15"`

33746

33747 - `"mcp-client-2026-09-15"`

33748

33749 - `"ce-plugins-2026-09-01"`

33750

33751 - `"spend-limit-reads-2026-09-26"`

33752

33753 - `"telemetry-destinations-2026-08-11"`

33754

33755 - `"webhooks-2026-10-07"`

33756

33757 - `"message-threads-2026-08-12"`

33758

33759 - `"analytics-usage-cost-2026-08-24"`

33760

33761 - `"experimental-claude-tag-2026-10-21"`

33762

33763#### Returns

33764

33765- `data: array of BetaClaudeTagSlackChannel`

33766

33767 Channels on this page, newest first.

33768

33769 - `type: "slack_channel"`

33770

33771 - `id: string`

33772

33773 The channel's identifier. Starts with `slch_`. It does not change for as long as the channel is set up, with one exception: a channel that is not configured gets a new one when claude.ai replaces its record, which happens only while the organization's read-only mode is off. A channel that is removed and set up again also has a new one. The old one keeps returning 404; look the channel up by its Slack identifiers for the new one.

33774

33775 - `created_at: string`

33776

33777 Time the channel was set up, by whichever route. It does not change for as long as `id` does not.

33778

33779 format: date-time

33780

33781 - `settings: BetaClaudeTagSettings`

33782

33783 What the channel itself stores or, with `include_inherited`, what applies in it. `sources` says which.

33784

33785 - `enabled: boolean or null`

33786

33787 Whether Claude is on. The nearest settings level that sets it decides.

33788

33789 - `environment_id: string or null`

33790

33791 The environment Claude's sessions run in, starting with `env_`, or the identifier of a self-hosted runner pool, starting with `ccpool_`. The nearest settings level that sets it decides.

33792

33793 - `guest_policy: BetaClaudeTagGuestPolicy or null`

33794

33795 What Claude does in a channel while a Slack guest is in it. The nearest settings level that sets it decides.

33796

33797 - `BetaClaudeTagGuestPolicyAllow object`

33798

33799 Claude answers in the channel.

33800

33801 - `type: "allow"`

33802

33803 - `BetaClaudeTagGuestPolicyRestrict object`

33804

33805 Claude does not answer in the channel.

33806

33807 - `type: "restrict"`

33808

33809 - `BetaClaudeTagGuestPolicyChannelOnly object`

33810

33811 Claude answers with only the Bundles attached to the channel itself.

33812

33813 - `type: "channel_only"`

33814

33815 - `instructions: string or null`

33816

33817 Instructions for Claude. Instructions add up: the text of every settings level that sets it is joined, broadest first.

33818

33819 - `model: string or null`

33820

33821 The model Claude uses: a model identifier or the name of a model line. The nearest settings level that sets it decides. When none does, Claude uses the product's default model.

33822

33823 - `slack: BetaClaudeTagSlackChannelIdentifiers`

33824

33825 Slack's own identifiers for the channel.

33826

33827 - `channel_id: string`

33828

33829 Slack's identifier for the channel, starting with `C` or `G`, such as `C0456EFGH`.

33830

33831 minLength: 2, maxLength: 32

33832

33833 - `workspace_id: string`

33834

33835 Slack's identifier for the workspace the channel is in, starting with `T`, such as `T0123ABCD`.

33836

33837 minLength: 2, maxLength: 32

33838

33839 - `sources: BetaClaudeTagSettingSources`

33840

33841 For each setting, the settings level its value in `settings` comes from.

33842

33843 - `enabled: BetaClaudeTagSettingSource or null`

33844

33845 The settings level `enabled` comes from.

33846

33847 - `BetaClaudeTagSlackChannelSettingSource object`

33848

33849 The value comes from the addressed Slack channel itself.

33850

33851 - `type: "slack_channel"`

33852

33853 - `slack_channel_id: string`

33854

33855 The channel's identifier, starting with `slch_`.

33856

33857 - `BetaClaudeTagSlackWorkspaceSettingSource object`

33858

33859 The value comes from a Slack workspace of the organization.

33860

33861 - `type: "slack_workspace"`

33862

33863 - `slack_workspace_id: string`

33864

33865 The workspace's identifier, starting with `slws_`.

33866

33867 - `BetaClaudeTagSlackParentChannelSettingSource object`

33868

33869 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

33870

33871 - `type: "slack_parent_channel"`

33872

33873 - `slack_channel_id: string`

33874

33875 The parent channel's identifier, starting with `slch_`.

33876

33877 - `BetaClaudeTagSlackDefaultSettingSource object`

33878

33879 The value comes from the organization's Slack defaults, which have no identifier.

33880

33881 - `type: "slack_default"`

33882

33883 - `BetaClaudeTagDefaultSettingSource object`

33884

33885 No settings level sets a value, and the built-in default applies.

33886

33887 - `type: "default"`

33888

33889 - `BetaClaudeTagSlackGridChannelSettingSource object`

33890

33891 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

33892

33893 - `type: "slack_grid_channel"`

33894

33895 - `slack: BetaClaudeTagSlackChannelIdentifiers`

33896

33897 Slack's own identifiers for the channel under that other workspace.

33898

33899 - `environment_id: BetaClaudeTagSettingSource or null`

33900

33901 The settings level `environment_id` comes from.

33902

33903 - `guest_policy: BetaClaudeTagSettingSource or null`

33904

33905 The settings level `guest_policy` comes from.

33906

33907 - `instructions: array of BetaClaudeTagSettingSource`

33908

33909 Every settings level that contributes text to `instructions`, broadest first, in the order their text is joined. Empty when none does.

33910

33911 - `BetaClaudeTagSlackChannelSettingSource object`

33912

33913 The value comes from the addressed Slack channel itself.

33914

33915 - `BetaClaudeTagSlackWorkspaceSettingSource object`

33916

33917 The value comes from a Slack workspace of the organization.

33918

33919 - `BetaClaudeTagSlackParentChannelSettingSource object`

33920

33921 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

33922

33923 - `BetaClaudeTagSlackDefaultSettingSource object`

33924

33925 The value comes from the organization's Slack defaults, which have no identifier.

33926

33927 - `BetaClaudeTagDefaultSettingSource object`

33928

33929 No settings level sets a value, and the built-in default applies.

33930

33931 - `BetaClaudeTagSlackGridChannelSettingSource object`

33932

33933 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

33934

33935 - `model: BetaClaudeTagSettingSource or null`

33936

33937 The settings level `model` comes from.

33938

33939 - `updated_at: string`

33940

33941 Time Claude Tag's record of the channel last changed. It moves on every change to a setting, and can also move when something no response holds changes, so a later value says to read the channel again, not that a setting changed. Attaching or detaching a Bundle does not by itself move it.

33942

33943 format: date-time

33944

33945- `next_page: string or null`

33946

33947 Cursor for the next page. Null on the last page.

33948

33949#### Example

33950

33951```bash

33952curl https://api.anthropic.com/v1/organizations/claude_tag/slack_channels \

33953 -H 'anthropic-version: 2026-10-21' \

33954 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

33955 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

33956```

33957

33958##### Response (200)

33959

33960```json

33961{

33962 "data": [

33963 {

33964 "id": "id",

33965 "created_at": "2019-12-27T18:11:19.117Z",

33966 "settings": {

33967 "enabled": true,

33968 "environment_id": "environment_id",

33969 "guest_policy": {

33970 "type": "allow"

33971 },

33972 "instructions": "instructions",

33973 "model": "model"

33974 },

33975 "slack": {

33976 "channel_id": "xx",

33977 "workspace_id": "xx"

33978 },

33979 "sources": {

33980 "enabled": {

33981 "slack_channel_id": "slack_channel_id",

33982 "type": "slack_channel"

33983 },

33984 "environment_id": {

33985 "slack_channel_id": "slack_channel_id",

33986 "type": "slack_channel"

33987 },

33988 "guest_policy": {

33989 "slack_channel_id": "slack_channel_id",

33990 "type": "slack_channel"

33991 },

33992 "instructions": [

33993 {

33994 "slack_channel_id": "slack_channel_id",

33995 "type": "slack_channel"

33996 }

33997 ],

33998 "model": {

33999 "slack_channel_id": "slack_channel_id",

34000 "type": "slack_channel"

34001 }

34002 },

34003 "type": "slack_channel",

34004 "updated_at": "2019-12-27T18:11:19.117Z"

34005 }

34006 ],

34007 "next_page": "next_page"

34008}

34009```

34010

34011### Get Slack Channel

34012

34013**GET** `/v1/organizations/claude_tag/slack_channels/{id}`

34014

34015Get one Slack channel and its settings. By default `settings` holds what the channel itself stores: a setting that is null there is not set on the channel and is inherited. With `include_inherited=true`, `settings` holds what applies in the channel once the settings levels above it are taken into account, and the request fails with 503 rather than resolve against a settings level that could not be read. Either way `sources` names the settings level each value comes from, so a caller tells the channel's own value from an inherited one by its source, not by whether it is null.

34016

34017#### Path parameters

34018

34019- `id: string`

34020

34021 The channel's identifier, starting with `slch_`.

34022

34023#### Query parameters

34024

34025- `include_inherited: optional boolean`

34026

34027 When `true`, `settings` holds what applies in the channel once the settings levels above it are taken into account, and every member of `sources` names the level a value comes from. When `false` or left out, `settings` holds what the channel itself stores.

34028

34029#### Headers

34030

34031- `"anthropic-version": optional string`

34032

34033 The version of the Claude API you want to use.

34034

34035 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

34036

34037- `"anthropic-beta": optional array of AnthropicBeta`

34038

34039 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

34040

34041 - `string`

34042

34043 - `"message-batches-2024-09-24"`

34044

34045 - `"prompt-caching-2024-07-31"`

34046

34047 - `"computer-use-2024-10-22"`

34048

34049 - `"computer-use-2025-01-24"`

34050

34051 - `"pdfs-2024-09-25"`

34052

34053 - `"token-counting-2024-11-01"`

34054

34055 - `"token-efficient-tools-2025-02-19"`

34056

34057 - `"output-128k-2025-02-19"`

34058

34059 - `"files-api-2025-04-14"`

34060

34061 - `"mcp-client-2025-04-04"`

34062

34063 - `"mcp-client-2025-11-20"`

34064

34065 - `"dev-full-thinking-2025-05-14"`

34066

34067 - `"interleaved-thinking-2025-05-14"`

34068

34069 - `"code-execution-2025-05-22"`

34070

34071 - `"extended-cache-ttl-2025-04-11"`

34072

34073 - `"context-1m-2025-08-07"`

34074

34075 - `"context-management-2025-06-27"`

34076

34077 - `"model-context-window-exceeded-2025-08-26"`

34078

34079 - `"skills-2025-10-02"`

34080

34081 - `"fast-mode-2026-02-01"`

34082

34083 - `"output-300k-2026-03-24"`

34084

34085 - `"user-profiles-2026-03-24"`

34086

34087 - `"user-profiles-2026-08-18"`

34088

34089 - `"user-profiles-2026-09-04"`

34090

34091 - `"advisor-tool-2026-03-01"`

34092

34093 - `"managed-agents-2026-04-01"`

34094

34095 - `"cache-diagnosis-2026-04-07"`

34096

34097 - `"dreaming-2026-04-21"`

34098

34099 - `"thinking-token-count-2026-05-13"`

34100

34101 - `"server-side-fallback-2026-06-01"`

34102

34103 - `"server-side-fallback-2026-07-01"`

34104

34105 - `"fallback-credit-2026-06-01"`

34106

34107 - `"fallback-credit-2026-07-01"`

34108

34109 - `"agent-memory-2026-07-22"`

34110

34111 - `"mid-conversation-tool-changes-2026-07-01"`

34112

34113 - `"compact-2026-01-12"`

34114

34115 - `"computer-use-2025-11-24"`

34116

34117 - `"mcp-tunnels-2026-06-22"`

34118

34119 - `"structured-outputs-2025-11-13"`

34120

34121 - `"task-budgets-2026-03-13"`

34122

34123 - `"thinking-display-updates-2026-08-18"`

34124

34125 - `"ce-user-management-2026-07-13"`

34126

34127 - `"mid-conversation-output-config-2026-07-01"`

34128

34129 - `"thinking-binding-controls-2026-08-01"`

34130

34131 - `"mid-conversation-system-clear-at-2026-08-21"`

34132

34133 - `"compact-2026-09-04"`

34134

34135 - `"inline-tools-2026-09-15"`

34136

34137 - `"mcp-client-2026-09-15"`

34138

34139 - `"ce-plugins-2026-09-01"`

34140

34141 - `"spend-limit-reads-2026-09-26"`

34142

34143 - `"telemetry-destinations-2026-08-11"`

34144

34145 - `"webhooks-2026-10-07"`

34146

34147 - `"message-threads-2026-08-12"`

34148

34149 - `"analytics-usage-cost-2026-08-24"`

34150

34151 - `"experimental-claude-tag-2026-10-21"`

34152

34153#### Returns

34154

34155- `BetaClaudeTagSlackChannel object`

34156

34157 A Slack channel set up for Claude Tag, with the settings it stores or, with `include_inherited`, the settings that apply in it.

34158

34159 - `type: "slack_channel"`

34160

34161 - `id: string`

34162

34163 The channel's identifier. Starts with `slch_`. It does not change for as long as the channel is set up, with one exception: a channel that is not configured gets a new one when claude.ai replaces its record, which happens only while the organization's read-only mode is off. A channel that is removed and set up again also has a new one. The old one keeps returning 404; look the channel up by its Slack identifiers for the new one.

34164

34165 - `created_at: string`

34166

34167 Time the channel was set up, by whichever route. It does not change for as long as `id` does not.

34168

34169 format: date-time

34170

34171 - `settings: BetaClaudeTagSettings`

34172

34173 What the channel itself stores or, with `include_inherited`, what applies in it. `sources` says which.

34174

34175 - `enabled: boolean or null`

34176

34177 Whether Claude is on. The nearest settings level that sets it decides.

34178

34179 - `environment_id: string or null`

34180

34181 The environment Claude's sessions run in, starting with `env_`, or the identifier of a self-hosted runner pool, starting with `ccpool_`. The nearest settings level that sets it decides.

34182

34183 - `guest_policy: BetaClaudeTagGuestPolicy or null`

34184

34185 What Claude does in a channel while a Slack guest is in it. The nearest settings level that sets it decides.

34186

34187 - `BetaClaudeTagGuestPolicyAllow object`

34188

34189 Claude answers in the channel.

34190

34191 - `type: "allow"`

34192

34193 - `BetaClaudeTagGuestPolicyRestrict object`

34194

34195 Claude does not answer in the channel.

34196

34197 - `type: "restrict"`

34198

34199 - `BetaClaudeTagGuestPolicyChannelOnly object`

34200

34201 Claude answers with only the Bundles attached to the channel itself.

34202

34203 - `type: "channel_only"`

34204

34205 - `instructions: string or null`

34206

34207 Instructions for Claude. Instructions add up: the text of every settings level that sets it is joined, broadest first.

34208

34209 - `model: string or null`

34210

34211 The model Claude uses: a model identifier or the name of a model line. The nearest settings level that sets it decides. When none does, Claude uses the product's default model.

34212

34213 - `slack: BetaClaudeTagSlackChannelIdentifiers`

34214

34215 Slack's own identifiers for the channel.

34216

34217 - `channel_id: string`

34218

34219 Slack's identifier for the channel, starting with `C` or `G`, such as `C0456EFGH`.

34220

34221 minLength: 2, maxLength: 32

34222

34223 - `workspace_id: string`

34224

34225 Slack's identifier for the workspace the channel is in, starting with `T`, such as `T0123ABCD`.

34226

34227 minLength: 2, maxLength: 32

34228

34229 - `sources: BetaClaudeTagSettingSources`

34230

34231 For each setting, the settings level its value in `settings` comes from.

34232

34233 - `enabled: BetaClaudeTagSettingSource or null`

34234

34235 The settings level `enabled` comes from.

34236

34237 - `BetaClaudeTagSlackChannelSettingSource object`

34238

34239 The value comes from the addressed Slack channel itself.

34240

34241 - `type: "slack_channel"`

34242

34243 - `slack_channel_id: string`

34244

34245 The channel's identifier, starting with `slch_`.

34246

34247 - `BetaClaudeTagSlackWorkspaceSettingSource object`

34248

34249 The value comes from a Slack workspace of the organization.

34250

34251 - `type: "slack_workspace"`

34252

34253 - `slack_workspace_id: string`

34254

34255 The workspace's identifier, starting with `slws_`.

34256

34257 - `BetaClaudeTagSlackParentChannelSettingSource object`

34258

34259 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

34260

34261 - `type: "slack_parent_channel"`

34262

34263 - `slack_channel_id: string`

34264

34265 The parent channel's identifier, starting with `slch_`.

34266

34267 - `BetaClaudeTagSlackDefaultSettingSource object`

34268

34269 The value comes from the organization's Slack defaults, which have no identifier.

34270

34271 - `type: "slack_default"`

34272

34273 - `BetaClaudeTagDefaultSettingSource object`

34274

34275 No settings level sets a value, and the built-in default applies.

34276

34277 - `type: "default"`

34278

34279 - `BetaClaudeTagSlackGridChannelSettingSource object`

34280

34281 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

34282

34283 - `type: "slack_grid_channel"`

34284

34285 - `slack: BetaClaudeTagSlackChannelIdentifiers`

34286

34287 Slack's own identifiers for the channel under that other workspace.

34288

34289 - `environment_id: BetaClaudeTagSettingSource or null`

34290

34291 The settings level `environment_id` comes from.

34292

34293 - `guest_policy: BetaClaudeTagSettingSource or null`

34294

34295 The settings level `guest_policy` comes from.

34296

34297 - `instructions: array of BetaClaudeTagSettingSource`

34298

34299 Every settings level that contributes text to `instructions`, broadest first, in the order their text is joined. Empty when none does.

34300

34301 - `BetaClaudeTagSlackChannelSettingSource object`

34302

34303 The value comes from the addressed Slack channel itself.

34304

34305 - `BetaClaudeTagSlackWorkspaceSettingSource object`

34306

34307 The value comes from a Slack workspace of the organization.

34308

34309 - `BetaClaudeTagSlackParentChannelSettingSource object`

34310

34311 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

34312

34313 - `BetaClaudeTagSlackDefaultSettingSource object`

34314

34315 The value comes from the organization's Slack defaults, which have no identifier.

34316

34317 - `BetaClaudeTagDefaultSettingSource object`

34318

34319 No settings level sets a value, and the built-in default applies.

34320

34321 - `BetaClaudeTagSlackGridChannelSettingSource object`

34322

34323 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

34324

34325 - `model: BetaClaudeTagSettingSource or null`

34326

34327 The settings level `model` comes from.

34328

34329 - `updated_at: string`

34330

34331 Time Claude Tag's record of the channel last changed. It moves on every change to a setting, and can also move when something no response holds changes, so a later value says to read the channel again, not that a setting changed. Attaching or detaching a Bundle does not by itself move it.

34332

34333 format: date-time

34334

34335#### Example

34336

34337```bash

34338curl https://api.anthropic.com/v1/organizations/claude_tag/slack_channels/$ID \

34339 -H 'anthropic-version: 2026-10-21' \

34340 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

34341 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

34342```

34343

34344##### Response (200)

34345

34346```json

34347{

34348 "id": "id",

34349 "created_at": "2019-12-27T18:11:19.117Z",

34350 "settings": {

34351 "enabled": true,

34352 "environment_id": "environment_id",

34353 "guest_policy": {

34354 "type": "allow"

34355 },

34356 "instructions": "instructions",

34357 "model": "model"

34358 },

34359 "slack": {

34360 "channel_id": "xx",

34361 "workspace_id": "xx"

34362 },

34363 "sources": {

34364 "enabled": {

34365 "slack_channel_id": "slack_channel_id",

34366 "type": "slack_channel"

34367 },

34368 "environment_id": {

34369 "slack_channel_id": "slack_channel_id",

34370 "type": "slack_channel"

34371 },

34372 "guest_policy": {

34373 "slack_channel_id": "slack_channel_id",

34374 "type": "slack_channel"

34375 },

34376 "instructions": [

34377 {

34378 "slack_channel_id": "slack_channel_id",

34379 "type": "slack_channel"

34380 }

34381 ],

34382 "model": {

34383 "slack_channel_id": "slack_channel_id",

34384 "type": "slack_channel"

34385 }

34386 },

34387 "type": "slack_channel",

34388 "updated_at": "2019-12-27T18:11:19.117Z"

34389}

34390```

34391

34392### Update Slack Channel

34393

34394**POST** `/v1/organizations/claude_tag/slack_channels/{id}`

34395

34396Change the settings a Slack channel stores. The update is partial: a setting left out is unchanged, and a setting sent as null is cleared, so the channel inherits it. If any value is refused, no setting in the request is stored. A group direct message takes no settings and returns 400.

34397

34398#### Path parameters

34399

34400- `id: string`

34401

34402 The channel's identifier, starting with `slch_`.

34403

34404#### Headers

34405

34406- `"anthropic-version": optional string`

34407

34408 The version of the Claude API you want to use.

34409

34410 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

34411

34412- `"anthropic-beta": optional array of AnthropicBeta`

34413

34414 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

34415

34416 - `string`

34417

34418 - `"message-batches-2024-09-24"`

34419

34420 - `"prompt-caching-2024-07-31"`

34421

34422 - `"computer-use-2024-10-22"`

34423

34424 - `"computer-use-2025-01-24"`

34425

34426 - `"pdfs-2024-09-25"`

34427

34428 - `"token-counting-2024-11-01"`

34429

34430 - `"token-efficient-tools-2025-02-19"`

34431

34432 - `"output-128k-2025-02-19"`

34433

34434 - `"files-api-2025-04-14"`

34435

34436 - `"mcp-client-2025-04-04"`

34437

34438 - `"mcp-client-2025-11-20"`

34439

34440 - `"dev-full-thinking-2025-05-14"`

34441

34442 - `"interleaved-thinking-2025-05-14"`

34443

34444 - `"code-execution-2025-05-22"`

34445

34446 - `"extended-cache-ttl-2025-04-11"`

34447

34448 - `"context-1m-2025-08-07"`

34449

34450 - `"context-management-2025-06-27"`

34451

34452 - `"model-context-window-exceeded-2025-08-26"`

34453

34454 - `"skills-2025-10-02"`

34455

34456 - `"fast-mode-2026-02-01"`

34457

34458 - `"output-300k-2026-03-24"`

34459

34460 - `"user-profiles-2026-03-24"`

34461

34462 - `"user-profiles-2026-08-18"`

34463

34464 - `"user-profiles-2026-09-04"`

34465

34466 - `"advisor-tool-2026-03-01"`

34467

34468 - `"managed-agents-2026-04-01"`

34469

34470 - `"cache-diagnosis-2026-04-07"`

34471

34472 - `"dreaming-2026-04-21"`

34473

34474 - `"thinking-token-count-2026-05-13"`

34475

34476 - `"server-side-fallback-2026-06-01"`

34477

34478 - `"server-side-fallback-2026-07-01"`

34479

34480 - `"fallback-credit-2026-06-01"`

34481

34482 - `"fallback-credit-2026-07-01"`

34483

34484 - `"agent-memory-2026-07-22"`

34485

34486 - `"mid-conversation-tool-changes-2026-07-01"`

34487

34488 - `"compact-2026-01-12"`

34489

34490 - `"computer-use-2025-11-24"`

34491

34492 - `"mcp-tunnels-2026-06-22"`

34493

34494 - `"structured-outputs-2025-11-13"`

34495

34496 - `"task-budgets-2026-03-13"`

34497

34498 - `"thinking-display-updates-2026-08-18"`

34499

34500 - `"ce-user-management-2026-07-13"`

34501

34502 - `"mid-conversation-output-config-2026-07-01"`

34503

34504 - `"thinking-binding-controls-2026-08-01"`

34505

34506 - `"mid-conversation-system-clear-at-2026-08-21"`

34507

34508 - `"compact-2026-09-04"`

34509

34510 - `"inline-tools-2026-09-15"`

34511

34512 - `"mcp-client-2026-09-15"`

34513

34514 - `"ce-plugins-2026-09-01"`

34515

34516 - `"spend-limit-reads-2026-09-26"`

34517

34518 - `"telemetry-destinations-2026-08-11"`

34519

34520 - `"webhooks-2026-10-07"`

34521

34522 - `"message-threads-2026-08-12"`

34523

34524 - `"analytics-usage-cost-2026-08-24"`

34525

34526 - `"experimental-claude-tag-2026-10-21"`

34527

34528#### Body parameters

34529

34530- `settings: BetaClaudeTagSettingsUpdate`

34531

34532 The settings to change.

34533

34534 - `enabled: optional boolean or null`

34535

34536 Whether Claude is on. `true` on a channel turns Claude on in it even when its workspace stores `false`. Null clears it.

34537

34538 - `environment_id: optional string or null`

34539

34540 The environment Claude's sessions run in: the identifier of an active environment of the organization, starting with `env_`. Null clears it.

34541

34542 minLength: 1, maxLength: 64

34543

34544 - `guest_policy: optional BetaClaudeTagGuestPolicy or null`

34545

34546 What Claude does in a channel while a Slack guest is in it. Null clears it.

34547

34548 - `BetaClaudeTagGuestPolicyAllow object`

34549

34550 Claude answers in the channel.

34551

34552 - `type: "allow"`

34553

34554 - `BetaClaudeTagGuestPolicyRestrict object`

34555

34556 Claude does not answer in the channel.

34557

34558 - `type: "restrict"`

34559

34560 - `BetaClaudeTagGuestPolicyChannelOnly object`

34561

34562 Claude answers with only the Bundles attached to the channel itself.

34563

34564 - `type: "channel_only"`

34565

34566 - `instructions: optional string or null`

34567

34568 Instructions for Claude, up to 8,192 characters after leading and trailing white space is trimmed. Null or blank text clears them.

34569

34570 - `model: optional string or null`

34571

34572 The model Claude uses: the identifier of a model, or the name of a model line, that the organization is entitled to for Claude Tag. Null clears it.

34573

34574 minLength: 1, maxLength: 128

34575

34576#### Returns

34577

34578- `BetaClaudeTagSlackChannel object`

34579

34580 A Slack channel set up for Claude Tag, with the settings it stores or, with `include_inherited`, the settings that apply in it.

34581

34582 - `type: "slack_channel"`

34583

34584 - `id: string`

34585

34586 The channel's identifier. Starts with `slch_`. It does not change for as long as the channel is set up, with one exception: a channel that is not configured gets a new one when claude.ai replaces its record, which happens only while the organization's read-only mode is off. A channel that is removed and set up again also has a new one. The old one keeps returning 404; look the channel up by its Slack identifiers for the new one.

34587

34588 - `created_at: string`

34589

34590 Time the channel was set up, by whichever route. It does not change for as long as `id` does not.

34591

34592 format: date-time

34593

34594 - `settings: BetaClaudeTagSettings`

34595

34596 What the channel itself stores or, with `include_inherited`, what applies in it. `sources` says which.

34597

34598 - `enabled: boolean or null`

34599

34600 Whether Claude is on. The nearest settings level that sets it decides.

34601

34602 - `environment_id: string or null`

34603

34604 The environment Claude's sessions run in, starting with `env_`, or the identifier of a self-hosted runner pool, starting with `ccpool_`. The nearest settings level that sets it decides.

34605

34606 - `guest_policy: BetaClaudeTagGuestPolicy or null`

34607

34608 What Claude does in a channel while a Slack guest is in it. The nearest settings level that sets it decides.

34609

34610 - `BetaClaudeTagGuestPolicyAllow object`

34611

34612 Claude answers in the channel.

34613

34614 - `type: "allow"`

34615

34616 - `BetaClaudeTagGuestPolicyRestrict object`

34617

34618 Claude does not answer in the channel.

34619

34620 - `type: "restrict"`

34621

34622 - `BetaClaudeTagGuestPolicyChannelOnly object`

34623

34624 Claude answers with only the Bundles attached to the channel itself.

34625

34626 - `type: "channel_only"`

34627

34628 - `instructions: string or null`

34629

34630 Instructions for Claude. Instructions add up: the text of every settings level that sets it is joined, broadest first.

34631

34632 - `model: string or null`

34633

34634 The model Claude uses: a model identifier or the name of a model line. The nearest settings level that sets it decides. When none does, Claude uses the product's default model.

34635

34636 - `slack: BetaClaudeTagSlackChannelIdentifiers`

34637

34638 Slack's own identifiers for the channel.

34639

34640 - `channel_id: string`

34641

34642 Slack's identifier for the channel, starting with `C` or `G`, such as `C0456EFGH`.

34643

34644 minLength: 2, maxLength: 32

34645

34646 - `workspace_id: string`

34647

34648 Slack's identifier for the workspace the channel is in, starting with `T`, such as `T0123ABCD`.

34649

34650 minLength: 2, maxLength: 32

34651

34652 - `sources: BetaClaudeTagSettingSources`

34653

34654 For each setting, the settings level its value in `settings` comes from.

34655

34656 - `enabled: BetaClaudeTagSettingSource or null`

34657

34658 The settings level `enabled` comes from.

34659

34660 - `BetaClaudeTagSlackChannelSettingSource object`

34661

34662 The value comes from the addressed Slack channel itself.

34663

34664 - `type: "slack_channel"`

34665

34666 - `slack_channel_id: string`

34667

34668 The channel's identifier, starting with `slch_`.

34669

34670 - `BetaClaudeTagSlackWorkspaceSettingSource object`

34671

34672 The value comes from a Slack workspace of the organization.

34673

34674 - `type: "slack_workspace"`

34675

34676 - `slack_workspace_id: string`

34677

34678 The workspace's identifier, starting with `slws_`.

34679

34680 - `BetaClaudeTagSlackParentChannelSettingSource object`

34681

34682 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

34683

34684 - `type: "slack_parent_channel"`

34685

34686 - `slack_channel_id: string`

34687

34688 The parent channel's identifier, starting with `slch_`.

34689

34690 - `BetaClaudeTagSlackDefaultSettingSource object`

34691

34692 The value comes from the organization's Slack defaults, which have no identifier.

34693

34694 - `type: "slack_default"`

34695

34696 - `BetaClaudeTagDefaultSettingSource object`

34697

34698 No settings level sets a value, and the built-in default applies.

34699

34700 - `type: "default"`

34701

34702 - `BetaClaudeTagSlackGridChannelSettingSource object`

34703

34704 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

34705

34706 - `type: "slack_grid_channel"`

34707

34708 - `slack: BetaClaudeTagSlackChannelIdentifiers`

34709

34710 Slack's own identifiers for the channel under that other workspace.

34711

34712 - `environment_id: BetaClaudeTagSettingSource or null`

34713

34714 The settings level `environment_id` comes from.

34715

34716 - `guest_policy: BetaClaudeTagSettingSource or null`

34717

34718 The settings level `guest_policy` comes from.

34719

34720 - `instructions: array of BetaClaudeTagSettingSource`

34721

34722 Every settings level that contributes text to `instructions`, broadest first, in the order their text is joined. Empty when none does.

34723

34724 - `BetaClaudeTagSlackChannelSettingSource object`

34725

34726 The value comes from the addressed Slack channel itself.

34727

34728 - `BetaClaudeTagSlackWorkspaceSettingSource object`

34729

34730 The value comes from a Slack workspace of the organization.

34731

34732 - `BetaClaudeTagSlackParentChannelSettingSource object`

34733

34734 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

34735

34736 - `BetaClaudeTagSlackDefaultSettingSource object`

34737

34738 The value comes from the organization's Slack defaults, which have no identifier.

34739

34740 - `BetaClaudeTagDefaultSettingSource object`

34741

34742 No settings level sets a value, and the built-in default applies.

34743

34744 - `BetaClaudeTagSlackGridChannelSettingSource object`

34745

34746 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

34747

34748 - `model: BetaClaudeTagSettingSource or null`

34749

34750 The settings level `model` comes from.

34751

34752 - `updated_at: string`

34753

34754 Time Claude Tag's record of the channel last changed. It moves on every change to a setting, and can also move when something no response holds changes, so a later value says to read the channel again, not that a setting changed. Attaching or detaching a Bundle does not by itself move it.

34755

34756 format: date-time

34757

34758#### Example

34759

34760```bash

34761curl https://api.anthropic.com/v1/organizations/claude_tag/slack_channels/$ID \

34762 -H 'Content-Type: application/json' \

34763 -H 'anthropic-version: 2026-10-21' \

34764 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

34765 -H "Authorization: Bearer $ANTHROPIC_API_KEY" \

34766 -d '{

34767 "settings": {}

34768 }'

34769```

34770

34771##### Response (200)

34772

34773```json

34774{

34775 "id": "id",

34776 "created_at": "2019-12-27T18:11:19.117Z",

34777 "settings": {

34778 "enabled": true,

34779 "environment_id": "environment_id",

34780 "guest_policy": {

34781 "type": "allow"

34782 },

34783 "instructions": "instructions",

34784 "model": "model"

34785 },

34786 "slack": {

34787 "channel_id": "xx",

34788 "workspace_id": "xx"

34789 },

34790 "sources": {

34791 "enabled": {

34792 "slack_channel_id": "slack_channel_id",

34793 "type": "slack_channel"

34794 },

34795 "environment_id": {

34796 "slack_channel_id": "slack_channel_id",

34797 "type": "slack_channel"

34798 },

34799 "guest_policy": {

34800 "slack_channel_id": "slack_channel_id",

34801 "type": "slack_channel"

34802 },

34803 "instructions": [

34804 {

34805 "slack_channel_id": "slack_channel_id",

34806 "type": "slack_channel"

34807 }

34808 ],

34809 "model": {

34810 "slack_channel_id": "slack_channel_id",

34811 "type": "slack_channel"

34812 }

34813 },

34814 "type": "slack_channel",

34815 "updated_at": "2019-12-27T18:11:19.117Z"

34816}

34817```

34818

34819### Remove Slack Channel

34820

34821**DELETE** `/v1/organizations/claude_tag/slack_channels/{id}`

34822

34823Remove a Slack channel from Claude Tag. This deletes the channel's stored settings and its Bundle Targets at once, and then, in the background, the channel's sessions and their transcripts, its memory, its routines and the artifacts published from it. None of it can be restored. The channel in Slack and its messages are unchanged. The request is refused with 400 and `error.details.error_code` `slack_channel_enabled` while `enabled` resolves to `true` for the channel, as a read with `include_inherited=true` returns it: set `enabled` to `false` on the channel first. This guards against removing a channel in use by mistake; it is not a way to keep Claude out, because the removal deletes the channel's own `enabled: false` with its other settings, after which the channel inherits again and Claude's next use of it can set it up anew with a new identifier. The identifier returns 404 as soon as the request returns, so a repeated removal returns 404. A group direct message returns 400, and so does a channel that holds configuration this API does not manage.

34824

34825#### Path parameters

34826

34827- `id: string`

34828

34829 The channel's identifier, starting with `slch_`.

34830

34831#### Headers

34832

34833- `"anthropic-version": optional string`

34834

34835 The version of the Claude API you want to use.

34836

34837 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

34838

34839- `"anthropic-beta": optional array of AnthropicBeta`

34840

34841 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

34842

34843 - `string`

34844

34845 - `"message-batches-2024-09-24"`

34846

34847 - `"prompt-caching-2024-07-31"`

34848

34849 - `"computer-use-2024-10-22"`

34850

34851 - `"computer-use-2025-01-24"`

34852

34853 - `"pdfs-2024-09-25"`

34854

34855 - `"token-counting-2024-11-01"`

34856

34857 - `"token-efficient-tools-2025-02-19"`

34858

34859 - `"output-128k-2025-02-19"`

34860

34861 - `"files-api-2025-04-14"`

34862

34863 - `"mcp-client-2025-04-04"`

34864

34865 - `"mcp-client-2025-11-20"`

34866

34867 - `"dev-full-thinking-2025-05-14"`

34868

34869 - `"interleaved-thinking-2025-05-14"`

34870

34871 - `"code-execution-2025-05-22"`

34872

34873 - `"extended-cache-ttl-2025-04-11"`

34874

34875 - `"context-1m-2025-08-07"`

34876

34877 - `"context-management-2025-06-27"`

34878

34879 - `"model-context-window-exceeded-2025-08-26"`

34880

34881 - `"skills-2025-10-02"`

34882

34883 - `"fast-mode-2026-02-01"`

34884

34885 - `"output-300k-2026-03-24"`

34886

34887 - `"user-profiles-2026-03-24"`

34888

34889 - `"user-profiles-2026-08-18"`

34890

34891 - `"user-profiles-2026-09-04"`

34892

34893 - `"advisor-tool-2026-03-01"`

34894

34895 - `"managed-agents-2026-04-01"`

34896

34897 - `"cache-diagnosis-2026-04-07"`

34898

34899 - `"dreaming-2026-04-21"`

34900

34901 - `"thinking-token-count-2026-05-13"`

34902

34903 - `"server-side-fallback-2026-06-01"`

34904

34905 - `"server-side-fallback-2026-07-01"`

34906

34907 - `"fallback-credit-2026-06-01"`

34908

34909 - `"fallback-credit-2026-07-01"`

34910

34911 - `"agent-memory-2026-07-22"`

34912

34913 - `"mid-conversation-tool-changes-2026-07-01"`

34914

34915 - `"compact-2026-01-12"`

34916

34917 - `"computer-use-2025-11-24"`

34918

34919 - `"mcp-tunnels-2026-06-22"`

34920

34921 - `"structured-outputs-2025-11-13"`

34922

34923 - `"task-budgets-2026-03-13"`

34924

34925 - `"thinking-display-updates-2026-08-18"`

34926

34927 - `"ce-user-management-2026-07-13"`

34928

34929 - `"mid-conversation-output-config-2026-07-01"`

34930

34931 - `"thinking-binding-controls-2026-08-01"`

34932

34933 - `"mid-conversation-system-clear-at-2026-08-21"`

34934

34935 - `"compact-2026-09-04"`

34936

34937 - `"inline-tools-2026-09-15"`

34938

34939 - `"mcp-client-2026-09-15"`

34940

34941 - `"ce-plugins-2026-09-01"`

34942

34943 - `"spend-limit-reads-2026-09-26"`

34944

34945 - `"telemetry-destinations-2026-08-11"`

34946

34947 - `"webhooks-2026-10-07"`

34948

34949 - `"message-threads-2026-08-12"`

34950

34951 - `"analytics-usage-cost-2026-08-24"`

34952

34953 - `"experimental-claude-tag-2026-10-21"`

34954

34955#### Returns

34956

34957- `BetaClaudeTagDeletedSlackChannel object`

34958

34959 Confirms that a Slack channel was removed from Claude Tag.

34960

34961 - `type: "slack_channel_deleted"`

34962

34963 - `id: string`

34964

34965 The removed channel's identifier.

34966

34967#### Example

34968

34969```bash

34970curl https://api.anthropic.com/v1/organizations/claude_tag/slack_channels/$ID \

34971 -X DELETE \

34972 -H 'anthropic-version: 2026-10-21' \

34973 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

34974 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

34975```

34976

34977##### Response (200)

34978

34979```json

34980{

34981 "id": "id",

34982 "type": "slack_channel_deleted"

34983}

34984```

34985

34986## Organization › Claude Tag › Slack Workspaces

34987

34988### List Slack Workspaces

34989

34990**GET** `/v1/organizations/claude_tag/slack_workspaces`

34991

34992List the set-up Slack workspaces the organization owns, newest first by `created_at`. Pass `slack_workspace_id` to look one workspace up by its Slack identifier: `data` then holds that workspace, or is empty when it is not set up. A workspace of a Slack Enterprise Grid the organization has connected returns 404 until it is set up. With `include_inherited=true` each workspace's `settings` holds what applies in it, and the request fails with 503 rather than resolve against a settings level that could not be read. A `page` cursor resumes only a request with the same `include_inherited`.

34993

34994#### Query parameters

34995

34996- `include_inherited: optional boolean`

34997

34998 When `true`, `settings` holds what applies in each workspace once the organization's Slack defaults are taken into account, and every member of `sources` names the level a value comes from. When `false` or left out, `settings` holds what each workspace itself stores.

34999

35000- `limit: optional number`

35001

35002 Number of items to return per page, between 1 and 100. Defaults to 20.

35003

35004 format: int32

35005

35006- `page: optional string`

35007

35008 Opaque cursor from a previous response's `next_page`.

35009

35010- `slack_workspace_id: optional string`

35011

35012 Slack's identifier for one workspace, as in the workspace object's `slack.workspace_id`, not its `id`. With it `data` holds that workspace alone, or nothing when it is not set up.

35013

35014#### Headers

35015

35016- `"anthropic-version": optional string`

35017

35018 The version of the Claude API you want to use.

35019

35020 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

35021

35022- `"anthropic-beta": optional array of AnthropicBeta`

35023

35024 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

35025

35026 - `string`

35027

35028 - `"message-batches-2024-09-24"`

35029

35030 - `"prompt-caching-2024-07-31"`

35031

35032 - `"computer-use-2024-10-22"`

35033

35034 - `"computer-use-2025-01-24"`

35035

35036 - `"pdfs-2024-09-25"`

35037

35038 - `"token-counting-2024-11-01"`

35039

35040 - `"token-efficient-tools-2025-02-19"`

35041

35042 - `"output-128k-2025-02-19"`

35043

35044 - `"files-api-2025-04-14"`

35045

35046 - `"mcp-client-2025-04-04"`

35047

35048 - `"mcp-client-2025-11-20"`

35049

35050 - `"dev-full-thinking-2025-05-14"`

35051

35052 - `"interleaved-thinking-2025-05-14"`

35053

35054 - `"code-execution-2025-05-22"`

35055

35056 - `"extended-cache-ttl-2025-04-11"`

35057

35058 - `"context-1m-2025-08-07"`

35059

35060 - `"context-management-2025-06-27"`

35061

35062 - `"model-context-window-exceeded-2025-08-26"`

35063

35064 - `"skills-2025-10-02"`

35065

35066 - `"fast-mode-2026-02-01"`

35067

35068 - `"output-300k-2026-03-24"`

35069

35070 - `"user-profiles-2026-03-24"`

35071

35072 - `"user-profiles-2026-08-18"`

35073

35074 - `"user-profiles-2026-09-04"`

35075

35076 - `"advisor-tool-2026-03-01"`

35077

35078 - `"managed-agents-2026-04-01"`

35079

35080 - `"cache-diagnosis-2026-04-07"`

35081

35082 - `"dreaming-2026-04-21"`

35083

35084 - `"thinking-token-count-2026-05-13"`

35085

35086 - `"server-side-fallback-2026-06-01"`

35087

35088 - `"server-side-fallback-2026-07-01"`

35089

35090 - `"fallback-credit-2026-06-01"`

35091

35092 - `"fallback-credit-2026-07-01"`

35093

35094 - `"agent-memory-2026-07-22"`

35095

35096 - `"mid-conversation-tool-changes-2026-07-01"`

35097

35098 - `"compact-2026-01-12"`

35099

35100 - `"computer-use-2025-11-24"`

35101

35102 - `"mcp-tunnels-2026-06-22"`

35103

35104 - `"structured-outputs-2025-11-13"`

35105

35106 - `"task-budgets-2026-03-13"`

35107

35108 - `"thinking-display-updates-2026-08-18"`

35109

35110 - `"ce-user-management-2026-07-13"`

35111

35112 - `"mid-conversation-output-config-2026-07-01"`

35113

35114 - `"thinking-binding-controls-2026-08-01"`

35115

35116 - `"mid-conversation-system-clear-at-2026-08-21"`

35117

35118 - `"compact-2026-09-04"`

35119

35120 - `"inline-tools-2026-09-15"`

35121

35122 - `"mcp-client-2026-09-15"`

35123

35124 - `"ce-plugins-2026-09-01"`

35125

35126 - `"spend-limit-reads-2026-09-26"`

35127

35128 - `"telemetry-destinations-2026-08-11"`

35129

35130 - `"webhooks-2026-10-07"`

35131

35132 - `"message-threads-2026-08-12"`

35133

35134 - `"analytics-usage-cost-2026-08-24"`

35135

35136 - `"experimental-claude-tag-2026-10-21"`

35137

35138#### Returns

35139

35140- `data: array of BetaClaudeTagSlackWorkspace`

35141

35142 Workspaces on this page, newest first.

35143

35144 - `type: "slack_workspace"`

35145

35146 - `id: string`

35147

35148 The workspace's identifier. Starts with `slws_`. It does not change for as long as the workspace is set up. A workspace that is removed and set up again has a new one, and the old one keeps returning 404; look the workspace up by its Slack identifier for the new one.

35149

35150 - `created_at: string`

35151

35152 Time the workspace was set up, by whichever route. It does not change for as long as `id` does not.

35153

35154 format: date-time

35155

35156 - `settings: BetaClaudeTagSettings`

35157

35158 What the workspace itself stores or, with `include_inherited`, what applies in it. `sources` says which.

35159

35160 - `enabled: boolean or null`

35161

35162 Whether Claude is on. The nearest settings level that sets it decides.

35163

35164 - `environment_id: string or null`

35165

35166 The environment Claude's sessions run in, starting with `env_`, or the identifier of a self-hosted runner pool, starting with `ccpool_`. The nearest settings level that sets it decides.

35167

35168 - `guest_policy: BetaClaudeTagGuestPolicy or null`

35169

35170 What Claude does in a channel while a Slack guest is in it. The nearest settings level that sets it decides.

35171

35172 - `BetaClaudeTagGuestPolicyAllow object`

35173

35174 Claude answers in the channel.

35175

35176 - `type: "allow"`

35177

35178 - `BetaClaudeTagGuestPolicyRestrict object`

35179

35180 Claude does not answer in the channel.

35181

35182 - `type: "restrict"`

35183

35184 - `BetaClaudeTagGuestPolicyChannelOnly object`

35185

35186 Claude answers with only the Bundles attached to the channel itself.

35187

35188 - `type: "channel_only"`

35189

35190 - `instructions: string or null`

35191

35192 Instructions for Claude. Instructions add up: the text of every settings level that sets it is joined, broadest first.

35193

35194 - `model: string or null`

35195

35196 The model Claude uses: a model identifier or the name of a model line. The nearest settings level that sets it decides. When none does, Claude uses the product's default model.

35197

35198 - `slack: BetaClaudeTagSlackWorkspaceIdentifiers`

35199

35200 Slack's own identifier for the workspace.

35201

35202 - `workspace_id: string`

35203

35204 Slack's identifier for the workspace, starting with `T`, such as `T0123ABCD`.

35205

35206 - `sources: BetaClaudeTagSettingSources`

35207

35208 For each setting, the settings level its value in `settings` comes from.

35209

35210 - `enabled: BetaClaudeTagSettingSource or null`

35211

35212 The settings level `enabled` comes from.

35213

35214 - `BetaClaudeTagSlackChannelSettingSource object`

35215

35216 The value comes from the addressed Slack channel itself.

35217

35218 - `type: "slack_channel"`

35219

35220 - `slack_channel_id: string`

35221

35222 The channel's identifier, starting with `slch_`.

35223

35224 - `BetaClaudeTagSlackWorkspaceSettingSource object`

35225

35226 The value comes from a Slack workspace of the organization.

35227

35228 - `type: "slack_workspace"`

35229

35230 - `slack_workspace_id: string`

35231

35232 The workspace's identifier, starting with `slws_`.

35233

35234 - `BetaClaudeTagSlackParentChannelSettingSource object`

35235

35236 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

35237

35238 - `type: "slack_parent_channel"`

35239

35240 - `slack_channel_id: string`

35241

35242 The parent channel's identifier, starting with `slch_`.

35243

35244 - `BetaClaudeTagSlackDefaultSettingSource object`

35245

35246 The value comes from the organization's Slack defaults, which have no identifier.

35247

35248 - `type: "slack_default"`

35249

35250 - `BetaClaudeTagDefaultSettingSource object`

35251

35252 No settings level sets a value, and the built-in default applies.

35253

35254 - `type: "default"`

35255

35256 - `BetaClaudeTagSlackGridChannelSettingSource object`

35257

35258 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

35259

35260 - `type: "slack_grid_channel"`

35261

35262 - `slack: BetaClaudeTagSlackChannelIdentifiers`

35263

35264 Slack's own identifiers for the channel under that other workspace.

35265

35266 - `channel_id: string`

35267

35268 Slack's identifier for the channel, starting with `C` or `G`, such as `C0456EFGH`.

35269

35270 minLength: 2, maxLength: 32

35271

35272 - `workspace_id: string`

35273

35274 Slack's identifier for the workspace the channel is in, starting with `T`, such as `T0123ABCD`.

35275

35276 minLength: 2, maxLength: 32

35277

35278 - `environment_id: BetaClaudeTagSettingSource or null`

35279

35280 The settings level `environment_id` comes from.

35281

35282 - `guest_policy: BetaClaudeTagSettingSource or null`

35283

35284 The settings level `guest_policy` comes from.

35285

35286 - `instructions: array of BetaClaudeTagSettingSource`

35287

35288 Every settings level that contributes text to `instructions`, broadest first, in the order their text is joined. Empty when none does.

35289

35290 - `BetaClaudeTagSlackChannelSettingSource object`

35291

35292 The value comes from the addressed Slack channel itself.

35293

35294 - `BetaClaudeTagSlackWorkspaceSettingSource object`

35295

35296 The value comes from a Slack workspace of the organization.

35297

35298 - `BetaClaudeTagSlackParentChannelSettingSource object`

35299

35300 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

35301

35302 - `BetaClaudeTagSlackDefaultSettingSource object`

35303

35304 The value comes from the organization's Slack defaults, which have no identifier.

35305

35306 - `BetaClaudeTagDefaultSettingSource object`

35307

35308 No settings level sets a value, and the built-in default applies.

35309

35310 - `BetaClaudeTagSlackGridChannelSettingSource object`

35311

35312 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

35313

35314 - `model: BetaClaudeTagSettingSource or null`

35315

35316 The settings level `model` comes from.

35317

35318 - `updated_at: string`

35319

35320 Time Claude Tag's record of the workspace last changed. It moves on every change to a setting, and can also move when something no response holds changes, so a later value says to read the workspace again, not that a setting changed. Attaching or detaching a Bundle does not by itself move it.

35321

35322 format: date-time

35323

35324- `next_page: string or null`

35325

35326 Cursor for the next page. Null on the last page.

35327

35328#### Example

35329

35330```bash

35331curl https://api.anthropic.com/v1/organizations/claude_tag/slack_workspaces \

35332 -H 'anthropic-version: 2026-10-21' \

35333 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

35334 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

35335```

35336

35337##### Response (200)

35338

35339```json

35340{

35341 "data": [

35342 {

35343 "id": "id",

35344 "created_at": "2019-12-27T18:11:19.117Z",

35345 "settings": {

35346 "enabled": true,

35347 "environment_id": "environment_id",

35348 "guest_policy": {

35349 "type": "allow"

35350 },

35351 "instructions": "instructions",

35352 "model": "model"

35353 },

35354 "slack": {

35355 "workspace_id": "workspace_id"

35356 },

35357 "sources": {

35358 "enabled": {

35359 "slack_channel_id": "slack_channel_id",

35360 "type": "slack_channel"

35361 },

35362 "environment_id": {

35363 "slack_channel_id": "slack_channel_id",

35364 "type": "slack_channel"

35365 },

35366 "guest_policy": {

35367 "slack_channel_id": "slack_channel_id",

35368 "type": "slack_channel"

35369 },

35370 "instructions": [

35371 {

35372 "slack_channel_id": "slack_channel_id",

35373 "type": "slack_channel"

35374 }

35375 ],

35376 "model": {

35377 "slack_channel_id": "slack_channel_id",

35378 "type": "slack_channel"

35379 }

35380 },

35381 "type": "slack_workspace",

35382 "updated_at": "2019-12-27T18:11:19.117Z"

35383 }

35384 ],

35385 "next_page": "next_page"

35386}

35387```

35388

35389### Get Slack Workspace

35390

35391**GET** `/v1/organizations/claude_tag/slack_workspaces/{id}`

35392

35393Get one Slack workspace and its settings. By default `settings` holds what the workspace itself stores: a setting that is null there is not set on the workspace and is inherited. With `include_inherited=true`, `settings` holds what applies in the workspace once the organization's Slack defaults are taken into account, and the request fails with 503 rather than resolve against a settings level that could not be read. Either way `sources` names the settings level each value comes from.

35394

35395#### Path parameters

35396

35397- `id: string`

35398

35399 The workspace's identifier, starting with `slws_`.

35400

35401#### Query parameters

35402

35403- `include_inherited: optional boolean`

35404

35405 When `true`, `settings` holds what applies in the workspace once the organization's Slack defaults are taken into account, and every member of `sources` names the level a value comes from. When `false` or left out, `settings` holds what the workspace itself stores.

35406

35407#### Headers

35408

35409- `"anthropic-version": optional string`

35410

35411 The version of the Claude API you want to use.

35412

35413 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

35414

35415- `"anthropic-beta": optional array of AnthropicBeta`

35416

35417 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

35418

35419 - `string`

35420

35421 - `"message-batches-2024-09-24"`

35422

35423 - `"prompt-caching-2024-07-31"`

35424

35425 - `"computer-use-2024-10-22"`

35426

35427 - `"computer-use-2025-01-24"`

35428

35429 - `"pdfs-2024-09-25"`

35430

35431 - `"token-counting-2024-11-01"`

35432

35433 - `"token-efficient-tools-2025-02-19"`

35434

35435 - `"output-128k-2025-02-19"`

35436

35437 - `"files-api-2025-04-14"`

35438

35439 - `"mcp-client-2025-04-04"`

35440

35441 - `"mcp-client-2025-11-20"`

35442

35443 - `"dev-full-thinking-2025-05-14"`

35444

35445 - `"interleaved-thinking-2025-05-14"`

35446

35447 - `"code-execution-2025-05-22"`

35448

35449 - `"extended-cache-ttl-2025-04-11"`

35450

35451 - `"context-1m-2025-08-07"`

35452

35453 - `"context-management-2025-06-27"`

35454

35455 - `"model-context-window-exceeded-2025-08-26"`

35456

35457 - `"skills-2025-10-02"`

35458

35459 - `"fast-mode-2026-02-01"`

35460

35461 - `"output-300k-2026-03-24"`

35462

35463 - `"user-profiles-2026-03-24"`

35464

35465 - `"user-profiles-2026-08-18"`

35466

35467 - `"user-profiles-2026-09-04"`

35468

35469 - `"advisor-tool-2026-03-01"`

35470

35471 - `"managed-agents-2026-04-01"`

35472

35473 - `"cache-diagnosis-2026-04-07"`

35474

35475 - `"dreaming-2026-04-21"`

35476

35477 - `"thinking-token-count-2026-05-13"`

35478

35479 - `"server-side-fallback-2026-06-01"`

35480

35481 - `"server-side-fallback-2026-07-01"`

35482

35483 - `"fallback-credit-2026-06-01"`

35484

35485 - `"fallback-credit-2026-07-01"`

35486

35487 - `"agent-memory-2026-07-22"`

35488

35489 - `"mid-conversation-tool-changes-2026-07-01"`

35490

35491 - `"compact-2026-01-12"`

35492

35493 - `"computer-use-2025-11-24"`

35494

35495 - `"mcp-tunnels-2026-06-22"`

35496

35497 - `"structured-outputs-2025-11-13"`

35498

35499 - `"task-budgets-2026-03-13"`

35500

35501 - `"thinking-display-updates-2026-08-18"`

35502

35503 - `"ce-user-management-2026-07-13"`

35504

35505 - `"mid-conversation-output-config-2026-07-01"`

35506

35507 - `"thinking-binding-controls-2026-08-01"`

35508

35509 - `"mid-conversation-system-clear-at-2026-08-21"`

35510

35511 - `"compact-2026-09-04"`

35512

35513 - `"inline-tools-2026-09-15"`

35514

35515 - `"mcp-client-2026-09-15"`

35516

35517 - `"ce-plugins-2026-09-01"`

35518

35519 - `"spend-limit-reads-2026-09-26"`

35520

35521 - `"telemetry-destinations-2026-08-11"`

35522

35523 - `"webhooks-2026-10-07"`

35524

35525 - `"message-threads-2026-08-12"`

35526

35527 - `"analytics-usage-cost-2026-08-24"`

35528

35529 - `"experimental-claude-tag-2026-10-21"`

35530

35531#### Returns

35532

35533- `BetaClaudeTagSlackWorkspace object`

35534

35535 A Slack workspace the organization has connected, itself or through its Slack Enterprise Grid, with the settings it stores or, with `include_inherited`, the settings that apply in it.

35536

35537 - `type: "slack_workspace"`

35538

35539 - `id: string`

35540

35541 The workspace's identifier. Starts with `slws_`. It does not change for as long as the workspace is set up. A workspace that is removed and set up again has a new one, and the old one keeps returning 404; look the workspace up by its Slack identifier for the new one.

35542

35543 - `created_at: string`

35544

35545 Time the workspace was set up, by whichever route. It does not change for as long as `id` does not.

35546

35547 format: date-time

35548

35549 - `settings: BetaClaudeTagSettings`

35550

35551 What the workspace itself stores or, with `include_inherited`, what applies in it. `sources` says which.

35552

35553 - `enabled: boolean or null`

35554

35555 Whether Claude is on. The nearest settings level that sets it decides.

35556

35557 - `environment_id: string or null`

35558

35559 The environment Claude's sessions run in, starting with `env_`, or the identifier of a self-hosted runner pool, starting with `ccpool_`. The nearest settings level that sets it decides.

35560

35561 - `guest_policy: BetaClaudeTagGuestPolicy or null`

35562

35563 What Claude does in a channel while a Slack guest is in it. The nearest settings level that sets it decides.

35564

35565 - `BetaClaudeTagGuestPolicyAllow object`

35566

35567 Claude answers in the channel.

35568

35569 - `type: "allow"`

35570

35571 - `BetaClaudeTagGuestPolicyRestrict object`

35572

35573 Claude does not answer in the channel.

35574

35575 - `type: "restrict"`

35576

35577 - `BetaClaudeTagGuestPolicyChannelOnly object`

35578

35579 Claude answers with only the Bundles attached to the channel itself.

35580

35581 - `type: "channel_only"`

35582

35583 - `instructions: string or null`

35584

35585 Instructions for Claude. Instructions add up: the text of every settings level that sets it is joined, broadest first.

35586

35587 - `model: string or null`

35588

35589 The model Claude uses: a model identifier or the name of a model line. The nearest settings level that sets it decides. When none does, Claude uses the product's default model.

35590

35591 - `slack: BetaClaudeTagSlackWorkspaceIdentifiers`

35592

35593 Slack's own identifier for the workspace.

35594

35595 - `workspace_id: string`

35596

35597 Slack's identifier for the workspace, starting with `T`, such as `T0123ABCD`.

35598

35599 - `sources: BetaClaudeTagSettingSources`

35600

35601 For each setting, the settings level its value in `settings` comes from.

35602

35603 - `enabled: BetaClaudeTagSettingSource or null`

35604

35605 The settings level `enabled` comes from.

35606

35607 - `BetaClaudeTagSlackChannelSettingSource object`

35608

35609 The value comes from the addressed Slack channel itself.

35610

35611 - `type: "slack_channel"`

35612

35613 - `slack_channel_id: string`

35614

35615 The channel's identifier, starting with `slch_`.

35616

35617 - `BetaClaudeTagSlackWorkspaceSettingSource object`

35618

35619 The value comes from a Slack workspace of the organization.

35620

35621 - `type: "slack_workspace"`

35622

35623 - `slack_workspace_id: string`

35624

35625 The workspace's identifier, starting with `slws_`.

35626

35627 - `BetaClaudeTagSlackParentChannelSettingSource object`

35628

35629 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

35630

35631 - `type: "slack_parent_channel"`

35632

35633 - `slack_channel_id: string`

35634

35635 The parent channel's identifier, starting with `slch_`.

35636

35637 - `BetaClaudeTagSlackDefaultSettingSource object`

35638

35639 The value comes from the organization's Slack defaults, which have no identifier.

35640

35641 - `type: "slack_default"`

35642

35643 - `BetaClaudeTagDefaultSettingSource object`

35644

35645 No settings level sets a value, and the built-in default applies.

35646

35647 - `type: "default"`

35648

35649 - `BetaClaudeTagSlackGridChannelSettingSource object`

35650

35651 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

35652

35653 - `type: "slack_grid_channel"`

35654

35655 - `slack: BetaClaudeTagSlackChannelIdentifiers`

35656

35657 Slack's own identifiers for the channel under that other workspace.

35658

35659 - `channel_id: string`

35660

35661 Slack's identifier for the channel, starting with `C` or `G`, such as `C0456EFGH`.

35662

35663 minLength: 2, maxLength: 32

35664

35665 - `workspace_id: string`

35666

35667 Slack's identifier for the workspace the channel is in, starting with `T`, such as `T0123ABCD`.

35668

35669 minLength: 2, maxLength: 32

35670

35671 - `environment_id: BetaClaudeTagSettingSource or null`

35672

35673 The settings level `environment_id` comes from.

35674

35675 - `guest_policy: BetaClaudeTagSettingSource or null`

35676

35677 The settings level `guest_policy` comes from.

35678

35679 - `instructions: array of BetaClaudeTagSettingSource`

35680

35681 Every settings level that contributes text to `instructions`, broadest first, in the order their text is joined. Empty when none does.

35682

35683 - `BetaClaudeTagSlackChannelSettingSource object`

35684

35685 The value comes from the addressed Slack channel itself.

35686

35687 - `BetaClaudeTagSlackWorkspaceSettingSource object`

35688

35689 The value comes from a Slack workspace of the organization.

35690

35691 - `BetaClaudeTagSlackParentChannelSettingSource object`

35692

35693 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

35694

35695 - `BetaClaudeTagSlackDefaultSettingSource object`

35696

35697 The value comes from the organization's Slack defaults, which have no identifier.

35698

35699 - `BetaClaudeTagDefaultSettingSource object`

35700

35701 No settings level sets a value, and the built-in default applies.

35702

35703 - `BetaClaudeTagSlackGridChannelSettingSource object`

35704

35705 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

35706

35707 - `model: BetaClaudeTagSettingSource or null`

35708

35709 The settings level `model` comes from.

35710

35711 - `updated_at: string`

35712

35713 Time Claude Tag's record of the workspace last changed. It moves on every change to a setting, and can also move when something no response holds changes, so a later value says to read the workspace again, not that a setting changed. Attaching or detaching a Bundle does not by itself move it.

35714

35715 format: date-time

35716

35717#### Example

35718

35719```bash

35720curl https://api.anthropic.com/v1/organizations/claude_tag/slack_workspaces/$ID \

35721 -H 'anthropic-version: 2026-10-21' \

35722 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

35723 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

35724```

35725

35726##### Response (200)

35727

35728```json

35729{

35730 "id": "id",

35731 "created_at": "2019-12-27T18:11:19.117Z",

35732 "settings": {

35733 "enabled": true,

35734 "environment_id": "environment_id",

35735 "guest_policy": {

35736 "type": "allow"

35737 },

35738 "instructions": "instructions",

35739 "model": "model"

35740 },

35741 "slack": {

35742 "workspace_id": "workspace_id"

35743 },

35744 "sources": {

35745 "enabled": {

35746 "slack_channel_id": "slack_channel_id",

35747 "type": "slack_channel"

35748 },

35749 "environment_id": {

35750 "slack_channel_id": "slack_channel_id",

35751 "type": "slack_channel"

35752 },

35753 "guest_policy": {

35754 "slack_channel_id": "slack_channel_id",

35755 "type": "slack_channel"

35756 },

35757 "instructions": [

35758 {

35759 "slack_channel_id": "slack_channel_id",

35760 "type": "slack_channel"

35761 }

35762 ],

35763 "model": {

35764 "slack_channel_id": "slack_channel_id",

35765 "type": "slack_channel"

35766 }

35767 },

35768 "type": "slack_workspace",

35769 "updated_at": "2019-12-27T18:11:19.117Z"

35770}

35771```

35772

35773### Update Slack Workspace

35774

35775**POST** `/v1/organizations/claude_tag/slack_workspaces/{id}`

35776

35777Change the settings a Slack workspace stores. The update is partial: a setting left out is unchanged, and a setting sent as null is cleared, so the workspace inherits it. If any value is refused, no setting in the request is stored.

35778

35779#### Path parameters

35780

35781- `id: string`

35782

35783 The workspace's identifier, starting with `slws_`.

35784

35785#### Headers

35786

35787- `"anthropic-version": optional string`

35788

35789 The version of the Claude API you want to use.

35790

35791 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

35792

35793- `"anthropic-beta": optional array of AnthropicBeta`

35794

35795 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

35796

35797 - `string`

35798

35799 - `"message-batches-2024-09-24"`

35800

35801 - `"prompt-caching-2024-07-31"`

35802

35803 - `"computer-use-2024-10-22"`

35804

35805 - `"computer-use-2025-01-24"`

35806

35807 - `"pdfs-2024-09-25"`

35808

35809 - `"token-counting-2024-11-01"`

35810

35811 - `"token-efficient-tools-2025-02-19"`

35812

35813 - `"output-128k-2025-02-19"`

35814

35815 - `"files-api-2025-04-14"`

35816

35817 - `"mcp-client-2025-04-04"`

35818

35819 - `"mcp-client-2025-11-20"`

35820

35821 - `"dev-full-thinking-2025-05-14"`

35822

35823 - `"interleaved-thinking-2025-05-14"`

35824

35825 - `"code-execution-2025-05-22"`

35826

35827 - `"extended-cache-ttl-2025-04-11"`

35828

35829 - `"context-1m-2025-08-07"`

35830

35831 - `"context-management-2025-06-27"`

35832

35833 - `"model-context-window-exceeded-2025-08-26"`

35834

35835 - `"skills-2025-10-02"`

35836

35837 - `"fast-mode-2026-02-01"`

35838

35839 - `"output-300k-2026-03-24"`

35840

35841 - `"user-profiles-2026-03-24"`

35842

35843 - `"user-profiles-2026-08-18"`

35844

35845 - `"user-profiles-2026-09-04"`

35846

35847 - `"advisor-tool-2026-03-01"`

35848

35849 - `"managed-agents-2026-04-01"`

35850

35851 - `"cache-diagnosis-2026-04-07"`

35852

35853 - `"dreaming-2026-04-21"`

35854

35855 - `"thinking-token-count-2026-05-13"`

35856

35857 - `"server-side-fallback-2026-06-01"`

35858

35859 - `"server-side-fallback-2026-07-01"`

35860

35861 - `"fallback-credit-2026-06-01"`

35862

35863 - `"fallback-credit-2026-07-01"`

35864

35865 - `"agent-memory-2026-07-22"`

35866

35867 - `"mid-conversation-tool-changes-2026-07-01"`

35868

35869 - `"compact-2026-01-12"`

35870

35871 - `"computer-use-2025-11-24"`

35872

35873 - `"mcp-tunnels-2026-06-22"`

35874

35875 - `"structured-outputs-2025-11-13"`

35876

35877 - `"task-budgets-2026-03-13"`

35878

35879 - `"thinking-display-updates-2026-08-18"`

35880

35881 - `"ce-user-management-2026-07-13"`

35882

35883 - `"mid-conversation-output-config-2026-07-01"`

35884

35885 - `"thinking-binding-controls-2026-08-01"`

35886

35887 - `"mid-conversation-system-clear-at-2026-08-21"`

35888

35889 - `"compact-2026-09-04"`

35890

35891 - `"inline-tools-2026-09-15"`

35892

35893 - `"mcp-client-2026-09-15"`

35894

35895 - `"ce-plugins-2026-09-01"`

35896

35897 - `"spend-limit-reads-2026-09-26"`

35898

35899 - `"telemetry-destinations-2026-08-11"`

35900

35901 - `"webhooks-2026-10-07"`

35902

35903 - `"message-threads-2026-08-12"`

35904

35905 - `"analytics-usage-cost-2026-08-24"`

35906

35907 - `"experimental-claude-tag-2026-10-21"`

35908

35909#### Body parameters

35910

35911- `settings: BetaClaudeTagSettingsUpdate`

35912

35913 The settings to change.

35914

35915 - `enabled: optional boolean or null`

35916

35917 Whether Claude is on. `true` on a channel turns Claude on in it even when its workspace stores `false`. Null clears it.

35918

35919 - `environment_id: optional string or null`

35920

35921 The environment Claude's sessions run in: the identifier of an active environment of the organization, starting with `env_`. Null clears it.

35922

35923 minLength: 1, maxLength: 64

35924

35925 - `guest_policy: optional BetaClaudeTagGuestPolicy or null`

35926

35927 What Claude does in a channel while a Slack guest is in it. Null clears it.

35928

35929 - `BetaClaudeTagGuestPolicyAllow object`

35930

35931 Claude answers in the channel.

35932

35933 - `type: "allow"`

35934

35935 - `BetaClaudeTagGuestPolicyRestrict object`

35936

35937 Claude does not answer in the channel.

35938

35939 - `type: "restrict"`

35940

35941 - `BetaClaudeTagGuestPolicyChannelOnly object`

35942

35943 Claude answers with only the Bundles attached to the channel itself.

35944

35945 - `type: "channel_only"`

35946

35947 - `instructions: optional string or null`

35948

35949 Instructions for Claude, up to 8,192 characters after leading and trailing white space is trimmed. Null or blank text clears them.

35950

35951 - `model: optional string or null`

35952

35953 The model Claude uses: the identifier of a model, or the name of a model line, that the organization is entitled to for Claude Tag. Null clears it.

35954

35955 minLength: 1, maxLength: 128

35956

35957#### Returns

35958

35959- `BetaClaudeTagSlackWorkspace object`

35960

35961 A Slack workspace the organization has connected, itself or through its Slack Enterprise Grid, with the settings it stores or, with `include_inherited`, the settings that apply in it.

35962

35963 - `type: "slack_workspace"`

35964

35965 - `id: string`

35966

35967 The workspace's identifier. Starts with `slws_`. It does not change for as long as the workspace is set up. A workspace that is removed and set up again has a new one, and the old one keeps returning 404; look the workspace up by its Slack identifier for the new one.

35968

35969 - `created_at: string`

35970

35971 Time the workspace was set up, by whichever route. It does not change for as long as `id` does not.

35972

35973 format: date-time

35974

35975 - `settings: BetaClaudeTagSettings`

35976

35977 What the workspace itself stores or, with `include_inherited`, what applies in it. `sources` says which.

35978

35979 - `enabled: boolean or null`

35980

35981 Whether Claude is on. The nearest settings level that sets it decides.

35982

35983 - `environment_id: string or null`

35984

35985 The environment Claude's sessions run in, starting with `env_`, or the identifier of a self-hosted runner pool, starting with `ccpool_`. The nearest settings level that sets it decides.

35986

35987 - `guest_policy: BetaClaudeTagGuestPolicy or null`

35988

35989 What Claude does in a channel while a Slack guest is in it. The nearest settings level that sets it decides.

35990

35991 - `BetaClaudeTagGuestPolicyAllow object`

35992

35993 Claude answers in the channel.

35994

35995 - `type: "allow"`

35996

35997 - `BetaClaudeTagGuestPolicyRestrict object`

35998

35999 Claude does not answer in the channel.

36000

36001 - `type: "restrict"`

36002

36003 - `BetaClaudeTagGuestPolicyChannelOnly object`

36004

36005 Claude answers with only the Bundles attached to the channel itself.

36006

36007 - `type: "channel_only"`

36008

36009 - `instructions: string or null`

36010

36011 Instructions for Claude. Instructions add up: the text of every settings level that sets it is joined, broadest first.

36012

36013 - `model: string or null`

36014

36015 The model Claude uses: a model identifier or the name of a model line. The nearest settings level that sets it decides. When none does, Claude uses the product's default model.

36016

36017 - `slack: BetaClaudeTagSlackWorkspaceIdentifiers`

36018

36019 Slack's own identifier for the workspace.

36020

36021 - `workspace_id: string`

36022

36023 Slack's identifier for the workspace, starting with `T`, such as `T0123ABCD`.

36024

36025 - `sources: BetaClaudeTagSettingSources`

36026

36027 For each setting, the settings level its value in `settings` comes from.

36028

36029 - `enabled: BetaClaudeTagSettingSource or null`

36030

36031 The settings level `enabled` comes from.

36032

36033 - `BetaClaudeTagSlackChannelSettingSource object`

36034

36035 The value comes from the addressed Slack channel itself.

36036

36037 - `type: "slack_channel"`

36038

36039 - `slack_channel_id: string`

36040

36041 The channel's identifier, starting with `slch_`.

36042

36043 - `BetaClaudeTagSlackWorkspaceSettingSource object`

36044

36045 The value comes from a Slack workspace of the organization.

36046

36047 - `type: "slack_workspace"`

36048

36049 - `slack_workspace_id: string`

36050

36051 The workspace's identifier, starting with `slws_`.

36052

36053 - `BetaClaudeTagSlackParentChannelSettingSource object`

36054

36055 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

36056

36057 - `type: "slack_parent_channel"`

36058

36059 - `slack_channel_id: string`

36060

36061 The parent channel's identifier, starting with `slch_`.

36062

36063 - `BetaClaudeTagSlackDefaultSettingSource object`

36064

36065 The value comes from the organization's Slack defaults, which have no identifier.

36066

36067 - `type: "slack_default"`

36068

36069 - `BetaClaudeTagDefaultSettingSource object`

36070

36071 No settings level sets a value, and the built-in default applies.

36072

36073 - `type: "default"`

36074

36075 - `BetaClaudeTagSlackGridChannelSettingSource object`

36076

36077 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

36078

36079 - `type: "slack_grid_channel"`

36080

36081 - `slack: BetaClaudeTagSlackChannelIdentifiers`

36082

36083 Slack's own identifiers for the channel under that other workspace.

36084

36085 - `channel_id: string`

36086

36087 Slack's identifier for the channel, starting with `C` or `G`, such as `C0456EFGH`.

36088

36089 minLength: 2, maxLength: 32

36090

36091 - `workspace_id: string`

36092

36093 Slack's identifier for the workspace the channel is in, starting with `T`, such as `T0123ABCD`.

36094

36095 minLength: 2, maxLength: 32

36096

36097 - `environment_id: BetaClaudeTagSettingSource or null`

36098

36099 The settings level `environment_id` comes from.

36100

36101 - `guest_policy: BetaClaudeTagSettingSource or null`

36102

36103 The settings level `guest_policy` comes from.

36104

36105 - `instructions: array of BetaClaudeTagSettingSource`

36106

36107 Every settings level that contributes text to `instructions`, broadest first, in the order their text is joined. Empty when none does.

36108

36109 - `BetaClaudeTagSlackChannelSettingSource object`

36110

36111 The value comes from the addressed Slack channel itself.

36112

36113 - `BetaClaudeTagSlackWorkspaceSettingSource object`

36114

36115 The value comes from a Slack workspace of the organization.

36116

36117 - `BetaClaudeTagSlackParentChannelSettingSource object`

36118

36119 The value comes from the channel that a code channel was started from. A parent channel contributes `instructions` and `environment_id` only.

36120

36121 - `BetaClaudeTagSlackDefaultSettingSource object`

36122

36123 The value comes from the organization's Slack defaults, which have no identifier.

36124

36125 - `BetaClaudeTagDefaultSettingSource object`

36126

36127 No settings level sets a value, and the built-in default applies.

36128

36129 - `BetaClaudeTagSlackGridChannelSettingSource object`

36130

36131 The value comes from the addressed channel as it is set up under another workspace of its Enterprise Grid: a channel shared across the Grid keeps one Slack channel identifier in every workspace that sees it.

36132

36133 - `model: BetaClaudeTagSettingSource or null`

36134

36135 The settings level `model` comes from.

36136

36137 - `updated_at: string`

36138

36139 Time Claude Tag's record of the workspace last changed. It moves on every change to a setting, and can also move when something no response holds changes, so a later value says to read the workspace again, not that a setting changed. Attaching or detaching a Bundle does not by itself move it.

36140

36141 format: date-time

36142

36143#### Example

36144

36145```bash

36146curl https://api.anthropic.com/v1/organizations/claude_tag/slack_workspaces/$ID \

36147 -H 'Content-Type: application/json' \

36148 -H 'anthropic-version: 2026-10-21' \

36149 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

36150 -H "Authorization: Bearer $ANTHROPIC_API_KEY" \

36151 -d '{

36152 "settings": {}

36153 }'

36154```

36155

36156##### Response (200)

36157

36158```json

36159{

36160 "id": "id",

36161 "created_at": "2019-12-27T18:11:19.117Z",

36162 "settings": {

36163 "enabled": true,

36164 "environment_id": "environment_id",

36165 "guest_policy": {

36166 "type": "allow"

36167 },

36168 "instructions": "instructions",

36169 "model": "model"

36170 },

36171 "slack": {

36172 "workspace_id": "workspace_id"

36173 },

36174 "sources": {

36175 "enabled": {

36176 "slack_channel_id": "slack_channel_id",

36177 "type": "slack_channel"

36178 },

36179 "environment_id": {

36180 "slack_channel_id": "slack_channel_id",

36181 "type": "slack_channel"

36182 },

36183 "guest_policy": {

36184 "slack_channel_id": "slack_channel_id",

36185 "type": "slack_channel"

36186 },

36187 "instructions": [

36188 {

36189 "slack_channel_id": "slack_channel_id",

36190 "type": "slack_channel"

36191 }

36192 ],

36193 "model": {

36194 "slack_channel_id": "slack_channel_id",

36195 "type": "slack_channel"

36196 }

36197 },

36198 "type": "slack_workspace",

36199 "updated_at": "2019-12-27T18:11:19.117Z"

36200}

36201```

36202

36203## Organization › Claude Tag › Bundles

36204

36205### List Bundles

36206

36207**GET** `/v1/organizations/claude_tag/bundles`

36208

36209List the organization's Bundles, newest first by `created_at`. A Bundle that Claude Tag manages for itself is not listed.

36210

36211#### Query parameters

36212

36213- `limit: optional number`

36214

36215 Number of items to return per page, between 1 and 100. Defaults to 20.

36216

36217 format: int32

36218

36219- `page: optional string`

36220

36221 Opaque cursor from a previous response's `next_page`.

36222

36223#### Headers

36224

36225- `"anthropic-version": optional string`

36226

36227 The version of the Claude API you want to use.

36228

36229 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

36230

36231- `"anthropic-beta": optional array of AnthropicBeta`

36232

36233 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

36234

36235 - `string`

36236

36237 - `"message-batches-2024-09-24"`

36238

36239 - `"prompt-caching-2024-07-31"`

36240

36241 - `"computer-use-2024-10-22"`

36242

36243 - `"computer-use-2025-01-24"`

36244

36245 - `"pdfs-2024-09-25"`

36246

36247 - `"token-counting-2024-11-01"`

36248

36249 - `"token-efficient-tools-2025-02-19"`

36250

36251 - `"output-128k-2025-02-19"`

36252

36253 - `"files-api-2025-04-14"`

36254

36255 - `"mcp-client-2025-04-04"`

36256

36257 - `"mcp-client-2025-11-20"`

36258

36259 - `"dev-full-thinking-2025-05-14"`

36260

36261 - `"interleaved-thinking-2025-05-14"`

36262

36263 - `"code-execution-2025-05-22"`

36264

36265 - `"extended-cache-ttl-2025-04-11"`

36266

36267 - `"context-1m-2025-08-07"`

36268

36269 - `"context-management-2025-06-27"`

36270

36271 - `"model-context-window-exceeded-2025-08-26"`

36272

36273 - `"skills-2025-10-02"`

36274

36275 - `"fast-mode-2026-02-01"`

36276

36277 - `"output-300k-2026-03-24"`

36278

36279 - `"user-profiles-2026-03-24"`

36280

36281 - `"user-profiles-2026-08-18"`

36282

36283 - `"user-profiles-2026-09-04"`

36284

36285 - `"advisor-tool-2026-03-01"`

36286

36287 - `"managed-agents-2026-04-01"`

36288

36289 - `"cache-diagnosis-2026-04-07"`

36290

36291 - `"dreaming-2026-04-21"`

36292

36293 - `"thinking-token-count-2026-05-13"`

36294

36295 - `"server-side-fallback-2026-06-01"`

36296

36297 - `"server-side-fallback-2026-07-01"`

36298

36299 - `"fallback-credit-2026-06-01"`

36300

36301 - `"fallback-credit-2026-07-01"`

36302

36303 - `"agent-memory-2026-07-22"`

36304

36305 - `"mid-conversation-tool-changes-2026-07-01"`

36306

36307 - `"compact-2026-01-12"`

36308

36309 - `"computer-use-2025-11-24"`

36310

36311 - `"mcp-tunnels-2026-06-22"`

36312

36313 - `"structured-outputs-2025-11-13"`

36314

36315 - `"task-budgets-2026-03-13"`

36316

36317 - `"thinking-display-updates-2026-08-18"`

36318

36319 - `"ce-user-management-2026-07-13"`

36320

36321 - `"mid-conversation-output-config-2026-07-01"`

36322

36323 - `"thinking-binding-controls-2026-08-01"`

36324

36325 - `"mid-conversation-system-clear-at-2026-08-21"`

36326

36327 - `"compact-2026-09-04"`

36328

36329 - `"inline-tools-2026-09-15"`

36330

36331 - `"mcp-client-2026-09-15"`

36332

36333 - `"ce-plugins-2026-09-01"`

36334

36335 - `"spend-limit-reads-2026-09-26"`

36336

36337 - `"telemetry-destinations-2026-08-11"`

36338

36339 - `"webhooks-2026-10-07"`

36340

36341 - `"message-threads-2026-08-12"`

36342

36343 - `"analytics-usage-cost-2026-08-24"`

36344

36345 - `"experimental-claude-tag-2026-10-21"`

36346

36347#### Returns

36348

36349- `data: array of BetaClaudeTagBundle`

36350

36351 Bundles on this page, newest first.

36352

36353 - `type: "bundle"`

36354

36355 - `id: string`

36356

36357 The Bundle's identifier. Starts with `capp_`.

36358

36359 - `created_at: string`

36360

36361 Time the Bundle was created.

36362

36363 format: date-time

36364

36365 - `description: string or null`

36366

36367 What the Bundle is for. Null when it has none. Read-only.

36368

36369 - `display_name: string`

36370

36371 The Bundle's name. Always set. Read-only.

36372

36373 - `updated_at: string`

36374

36375 Time of the last change to the Bundle.

36376

36377 format: date-time

36378

36379- `next_page: string or null`

36380

36381 Cursor for the next page. Null on the last page.

36382

36383#### Example

36384

36385```bash

36386curl https://api.anthropic.com/v1/organizations/claude_tag/bundles \

36387 -H 'anthropic-version: 2026-10-21' \

36388 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

36389 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

36390```

36391

36392##### Response (200)

36393

36394```json

36395{

36396 "data": [

36397 {

36398 "id": "id",

36399 "created_at": "2019-12-27T18:11:19.117Z",

36400 "description": "description",

36401 "display_name": "display_name",

36402 "type": "bundle",

36403 "updated_at": "2019-12-27T18:11:19.117Z"

36404 }

36405 ],

36406 "next_page": "next_page"

36407}

36408```

36409

36410### Get Bundle

36411

36412**GET** `/v1/organizations/claude_tag/bundles/{bundle_id}`

36413

36414Get one Bundle. What the Bundle grants (connectors, repositories and credentials) is never returned.

36415

36416#### Path parameters

36417

36418- `bundle_id: string`

36419

36420 The Bundle's identifier, starting with `capp_`.

36421

36422#### Headers

36423

36424- `"anthropic-version": optional string`

36425

36426 The version of the Claude API you want to use.

36427

36428 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

36429

36430- `"anthropic-beta": optional array of AnthropicBeta`

36431

36432 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

36433

36434 - `string`

36435

36436 - `"message-batches-2024-09-24"`

36437

36438 - `"prompt-caching-2024-07-31"`

36439

36440 - `"computer-use-2024-10-22"`

36441

36442 - `"computer-use-2025-01-24"`

36443

36444 - `"pdfs-2024-09-25"`

36445

36446 - `"token-counting-2024-11-01"`

36447

36448 - `"token-efficient-tools-2025-02-19"`

36449

36450 - `"output-128k-2025-02-19"`

36451

36452 - `"files-api-2025-04-14"`

36453

36454 - `"mcp-client-2025-04-04"`

36455

36456 - `"mcp-client-2025-11-20"`

36457

36458 - `"dev-full-thinking-2025-05-14"`

36459

36460 - `"interleaved-thinking-2025-05-14"`

36461

36462 - `"code-execution-2025-05-22"`

36463

36464 - `"extended-cache-ttl-2025-04-11"`

36465

36466 - `"context-1m-2025-08-07"`

36467

36468 - `"context-management-2025-06-27"`

36469

36470 - `"model-context-window-exceeded-2025-08-26"`

36471

36472 - `"skills-2025-10-02"`

36473

36474 - `"fast-mode-2026-02-01"`

36475

36476 - `"output-300k-2026-03-24"`

36477

36478 - `"user-profiles-2026-03-24"`

36479

36480 - `"user-profiles-2026-08-18"`

36481

36482 - `"user-profiles-2026-09-04"`

36483

36484 - `"advisor-tool-2026-03-01"`

36485

36486 - `"managed-agents-2026-04-01"`

36487

36488 - `"cache-diagnosis-2026-04-07"`

36489

36490 - `"dreaming-2026-04-21"`

36491

36492 - `"thinking-token-count-2026-05-13"`

36493

36494 - `"server-side-fallback-2026-06-01"`

36495

36496 - `"server-side-fallback-2026-07-01"`

36497

36498 - `"fallback-credit-2026-06-01"`

36499

36500 - `"fallback-credit-2026-07-01"`

36501

36502 - `"agent-memory-2026-07-22"`

36503

36504 - `"mid-conversation-tool-changes-2026-07-01"`

36505

36506 - `"compact-2026-01-12"`

36507

36508 - `"computer-use-2025-11-24"`

36509

36510 - `"mcp-tunnels-2026-06-22"`

36511

36512 - `"structured-outputs-2025-11-13"`

36513

36514 - `"task-budgets-2026-03-13"`

36515

36516 - `"thinking-display-updates-2026-08-18"`

36517

36518 - `"ce-user-management-2026-07-13"`

36519

36520 - `"mid-conversation-output-config-2026-07-01"`

36521

36522 - `"thinking-binding-controls-2026-08-01"`

36523

36524 - `"mid-conversation-system-clear-at-2026-08-21"`

36525

36526 - `"compact-2026-09-04"`

36527

36528 - `"inline-tools-2026-09-15"`

36529

36530 - `"mcp-client-2026-09-15"`

36531

36532 - `"ce-plugins-2026-09-01"`

36533

36534 - `"spend-limit-reads-2026-09-26"`

36535

36536 - `"telemetry-destinations-2026-08-11"`

36537

36538 - `"webhooks-2026-10-07"`

36539

36540 - `"message-threads-2026-08-12"`

36541

36542 - `"analytics-usage-cost-2026-08-24"`

36543

36544 - `"experimental-claude-tag-2026-10-21"`

36545

36546#### Returns

36547

36548- `BetaClaudeTagBundle object`

36549

36550 A named set of access that Claude can use in a Slack channel or a Slack workspace: connectors, repositories and credentials. Created and edited in claude.ai.

36551

36552 - `type: "bundle"`

36553

36554 - `id: string`

36555

36556 The Bundle's identifier. Starts with `capp_`.

36557

36558 - `created_at: string`

36559

36560 Time the Bundle was created.

36561

36562 format: date-time

36563

36564 - `description: string or null`

36565

36566 What the Bundle is for. Null when it has none. Read-only.

36567

36568 - `display_name: string`

36569

36570 The Bundle's name. Always set. Read-only.

36571

36572 - `updated_at: string`

36573

36574 Time of the last change to the Bundle.

36575

36576 format: date-time

36577

36578#### Example

36579

36580```bash

36581curl https://api.anthropic.com/v1/organizations/claude_tag/bundles/$BUNDLE_ID \

36582 -H 'anthropic-version: 2026-10-21' \

36583 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

36584 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

36585```

36586

36587##### Response (200)

36588

36589```json

36590{

36591 "id": "id",

36592 "created_at": "2019-12-27T18:11:19.117Z",

36593 "description": "description",

36594 "display_name": "display_name",

36595 "type": "bundle",

36596 "updated_at": "2019-12-27T18:11:19.117Z"

36597}

36598```

36599

36600## Organization › Claude Tag › Bundles › Targets

36601

36602### List Bundle Targets

36603

36604**GET** `/v1/organizations/claude_tag/bundles/{bundle_id}/targets`

36605

36606List the Slack channels and Slack workspaces a Bundle is attached to, newest first by `created_at`. Only channels and workspaces the organization owns are listed; a Bundle that is attached to anything else has no Bundle Target for it.

36607

36608#### Path parameters

36609

36610- `bundle_id: string`

36611

36612 The Bundle's identifier, starting with `capp_`.

36613

36614#### Query parameters

36615

36616- `limit: optional number`

36617

36618 Number of items to return per page, between 1 and 100. Defaults to 20.

36619

36620 format: int32

36621

36622- `page: optional string`

36623

36624 Opaque cursor from a previous response's `next_page`.

36625

36626#### Headers

36627

36628- `"anthropic-version": optional string`

36629

36630 The version of the Claude API you want to use.

36631

36632 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

36633

36634- `"anthropic-beta": optional array of AnthropicBeta`

36635

36636 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

36637

36638 - `string`

36639

36640 - `"message-batches-2024-09-24"`

36641

36642 - `"prompt-caching-2024-07-31"`

36643

36644 - `"computer-use-2024-10-22"`

36645

36646 - `"computer-use-2025-01-24"`

36647

36648 - `"pdfs-2024-09-25"`

36649

36650 - `"token-counting-2024-11-01"`

36651

36652 - `"token-efficient-tools-2025-02-19"`

36653

36654 - `"output-128k-2025-02-19"`

36655

36656 - `"files-api-2025-04-14"`

36657

36658 - `"mcp-client-2025-04-04"`

36659

36660 - `"mcp-client-2025-11-20"`

36661

36662 - `"dev-full-thinking-2025-05-14"`

36663

36664 - `"interleaved-thinking-2025-05-14"`

36665

36666 - `"code-execution-2025-05-22"`

36667

36668 - `"extended-cache-ttl-2025-04-11"`

36669

36670 - `"context-1m-2025-08-07"`

36671

36672 - `"context-management-2025-06-27"`

36673

36674 - `"model-context-window-exceeded-2025-08-26"`

36675

36676 - `"skills-2025-10-02"`

36677

36678 - `"fast-mode-2026-02-01"`

36679

36680 - `"output-300k-2026-03-24"`

36681

36682 - `"user-profiles-2026-03-24"`

36683

36684 - `"user-profiles-2026-08-18"`

36685

36686 - `"user-profiles-2026-09-04"`

36687

36688 - `"advisor-tool-2026-03-01"`

36689

36690 - `"managed-agents-2026-04-01"`

36691

36692 - `"cache-diagnosis-2026-04-07"`

36693

36694 - `"dreaming-2026-04-21"`

36695

36696 - `"thinking-token-count-2026-05-13"`

36697

36698 - `"server-side-fallback-2026-06-01"`

36699

36700 - `"server-side-fallback-2026-07-01"`

36701

36702 - `"fallback-credit-2026-06-01"`

36703

36704 - `"fallback-credit-2026-07-01"`

36705

36706 - `"agent-memory-2026-07-22"`

36707

36708 - `"mid-conversation-tool-changes-2026-07-01"`

36709

36710 - `"compact-2026-01-12"`

36711

36712 - `"computer-use-2025-11-24"`

36713

36714 - `"mcp-tunnels-2026-06-22"`

36715

36716 - `"structured-outputs-2025-11-13"`

36717

36718 - `"task-budgets-2026-03-13"`

36719

36720 - `"thinking-display-updates-2026-08-18"`

36721

36722 - `"ce-user-management-2026-07-13"`

36723

36724 - `"mid-conversation-output-config-2026-07-01"`

36725

36726 - `"thinking-binding-controls-2026-08-01"`

36727

36728 - `"mid-conversation-system-clear-at-2026-08-21"`

36729

36730 - `"compact-2026-09-04"`

36731

36732 - `"inline-tools-2026-09-15"`

36733

36734 - `"mcp-client-2026-09-15"`

36735

36736 - `"ce-plugins-2026-09-01"`

36737

36738 - `"spend-limit-reads-2026-09-26"`

36739

36740 - `"telemetry-destinations-2026-08-11"`

36741

36742 - `"webhooks-2026-10-07"`

36743

36744 - `"message-threads-2026-08-12"`

36745

36746 - `"analytics-usage-cost-2026-08-24"`

36747

36748 - `"experimental-claude-tag-2026-10-21"`

36749

36750#### Returns

36751

36752- `data: array of BetaClaudeTagBundleTarget`

36753

36754 Bundle Targets on this page, newest first.

36755

36756 - `BetaClaudeTagSlackChannelBundleTarget object`

36757

36758 One Bundle attached to one Slack channel. The Bundle can be used in that channel only.

36759

36760 - `type: "slack_channel"`

36761

36762 - `id: string`

36763

36764 The Bundle Target's identifier. Starts with `btgt_`. No two Bundle Targets have the same one, and the same Bundle and target always give the same one, also after a detach and a new attach. It is opaque.

36765

36766 - `created_at: string`

36767

36768 Time the Bundle was attached.

36769

36770 format: date-time

36771

36772 - `slack_channel_id: string`

36773

36774 The identifier of the Slack channel the Bundle is attached to, starting with `slch_`. It is the channel object's `id`, not Slack's `channel_id`.

36775

36776 - `BetaClaudeTagSlackWorkspaceBundleTarget object`

36777

36778 One Bundle attached to one Slack workspace. The Bundle can be used in the workspace and in every channel in it.

36779

36780 - `type: "slack_workspace"`

36781

36782 - `id: string`

36783

36784 The Bundle Target's identifier. Starts with `btgt_`. No two Bundle Targets have the same one, and the same Bundle and target always give the same one, also after a detach and a new attach. It is opaque.

36785

36786 - `created_at: string`

36787

36788 Time the Bundle was attached.

36789

36790 format: date-time

36791

36792 - `slack_workspace_id: string`

36793

36794 The identifier of the Slack workspace the Bundle is attached to, starting with `slws_`.

36795

36796- `next_page: string or null`

36797

36798 Cursor for the next page. Null on the last page.

36799

36800#### Example

36801

36802```bash

36803curl https://api.anthropic.com/v1/organizations/claude_tag/bundles/$BUNDLE_ID/targets \

36804 -H 'anthropic-version: 2026-10-21' \

36805 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

36806 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

36807```

36808

36809##### Response (200)

36810

36811```json

36812{

36813 "data": [

36814 {

36815 "id": "id",

36816 "created_at": "2019-12-27T18:11:19.117Z",

36817 "slack_channel_id": "slack_channel_id",

36818 "type": "slack_channel"

36819 }

36820 ],

36821 "next_page": "next_page"

36822}

36823```

36824

36825### Attach Bundle

36826

36827**POST** `/v1/organizations/claude_tag/bundles/{bundle_id}/targets`

36828

36829Attach a Bundle to a Slack channel or a Slack workspace. A Bundle attached to a workspace can be used in the workspace and in every channel in it. Attaching a Bundle to a channel or workspace it is already attached to returns the existing Bundle Target. The channel or workspace must be set up: attaching never sets a channel up. A group direct message takes no Bundle and returns 400.

36830

36831#### Path parameters

36832

36833- `bundle_id: string`

36834

36835 The Bundle's identifier, starting with `capp_`.

36836

36837#### Headers

36838

36839- `"anthropic-version": optional string`

36840

36841 The version of the Claude API you want to use.

36842

36843 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

36844

36845- `"anthropic-beta": optional array of AnthropicBeta`

36846

36847 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

36848

36849 - `string`

36850

36851 - `"message-batches-2024-09-24"`

36852

36853 - `"prompt-caching-2024-07-31"`

36854

36855 - `"computer-use-2024-10-22"`

36856

36857 - `"computer-use-2025-01-24"`

36858

36859 - `"pdfs-2024-09-25"`

36860

36861 - `"token-counting-2024-11-01"`

36862

36863 - `"token-efficient-tools-2025-02-19"`

36864

36865 - `"output-128k-2025-02-19"`

36866

36867 - `"files-api-2025-04-14"`

36868

36869 - `"mcp-client-2025-04-04"`

36870

36871 - `"mcp-client-2025-11-20"`

36872

36873 - `"dev-full-thinking-2025-05-14"`

36874

36875 - `"interleaved-thinking-2025-05-14"`

36876

36877 - `"code-execution-2025-05-22"`

36878

36879 - `"extended-cache-ttl-2025-04-11"`

36880

36881 - `"context-1m-2025-08-07"`

36882

36883 - `"context-management-2025-06-27"`

36884

36885 - `"model-context-window-exceeded-2025-08-26"`

36886

36887 - `"skills-2025-10-02"`

36888

36889 - `"fast-mode-2026-02-01"`

36890

36891 - `"output-300k-2026-03-24"`

36892

36893 - `"user-profiles-2026-03-24"`

36894

36895 - `"user-profiles-2026-08-18"`

36896

36897 - `"user-profiles-2026-09-04"`

36898

36899 - `"advisor-tool-2026-03-01"`

36900

36901 - `"managed-agents-2026-04-01"`

36902

36903 - `"cache-diagnosis-2026-04-07"`

36904

36905 - `"dreaming-2026-04-21"`

36906

36907 - `"thinking-token-count-2026-05-13"`

36908

36909 - `"server-side-fallback-2026-06-01"`

36910

36911 - `"server-side-fallback-2026-07-01"`

36912

36913 - `"fallback-credit-2026-06-01"`

36914

36915 - `"fallback-credit-2026-07-01"`

36916

36917 - `"agent-memory-2026-07-22"`

36918

36919 - `"mid-conversation-tool-changes-2026-07-01"`

36920

36921 - `"compact-2026-01-12"`

36922

36923 - `"computer-use-2025-11-24"`

36924

36925 - `"mcp-tunnels-2026-06-22"`

36926

36927 - `"structured-outputs-2025-11-13"`

36928

36929 - `"task-budgets-2026-03-13"`

36930

36931 - `"thinking-display-updates-2026-08-18"`

36932

36933 - `"ce-user-management-2026-07-13"`

36934

36935 - `"mid-conversation-output-config-2026-07-01"`

36936

36937 - `"thinking-binding-controls-2026-08-01"`

36938

36939 - `"mid-conversation-system-clear-at-2026-08-21"`

36940

36941 - `"compact-2026-09-04"`

36942

36943 - `"inline-tools-2026-09-15"`

36944

36945 - `"mcp-client-2026-09-15"`

36946

36947 - `"ce-plugins-2026-09-01"`

36948

36949 - `"spend-limit-reads-2026-09-26"`

36950

36951 - `"telemetry-destinations-2026-08-11"`

36952

36953 - `"webhooks-2026-10-07"`

36954

36955 - `"message-threads-2026-08-12"`

36956

36957 - `"analytics-usage-cost-2026-08-24"`

36958

36959 - `"experimental-claude-tag-2026-10-21"`

36960

36961#### Body parameters

36962

36963- `target: BetaClaudeTagBundleTargetCreate`

36964

36965 - `BetaClaudeTagSlackChannelBundleTargetCreate object`

36966

36967 Attach the Bundle to a Slack channel.

36968

36969 - `type: "slack_channel"`

36970

36971 - `slack_channel_id: string`

36972

36973 The identifier of the Slack channel to attach the Bundle to, starting with `slch_`. It is the channel object's `id`, not Slack's `channel_id`.

36974

36975 minLength: 8, maxLength: 64

36976

36977 - `BetaClaudeTagSlackWorkspaceBundleTargetCreate object`

36978

36979 Attach the Bundle to a Slack workspace.

36980

36981 - `type: "slack_workspace"`

36982

36983 - `slack_workspace_id: string`

36984

36985 The identifier of the Slack workspace to attach the Bundle to, starting with `slws_`.

36986

36987 minLength: 8, maxLength: 64

36988

36989#### Returns

36990

36991- `BetaClaudeTagBundleTarget = BetaClaudeTagSlackChannelBundleTarget or BetaClaudeTagSlackWorkspaceBundleTarget`

36992

36993 One Bundle attached to one Slack channel or one Slack workspace. More types of target can be added.

36994

36995 - `BetaClaudeTagSlackChannelBundleTarget object`

36996

36997 One Bundle attached to one Slack channel. The Bundle can be used in that channel only.

36998

36999 - `type: "slack_channel"`

37000

37001 - `id: string`

37002

37003 The Bundle Target's identifier. Starts with `btgt_`. No two Bundle Targets have the same one, and the same Bundle and target always give the same one, also after a detach and a new attach. It is opaque.

37004

37005 - `created_at: string`

37006

37007 Time the Bundle was attached.

37008

37009 format: date-time

37010

37011 - `slack_channel_id: string`

37012

37013 The identifier of the Slack channel the Bundle is attached to, starting with `slch_`. It is the channel object's `id`, not Slack's `channel_id`.

37014

37015 - `BetaClaudeTagSlackWorkspaceBundleTarget object`

37016

37017 One Bundle attached to one Slack workspace. The Bundle can be used in the workspace and in every channel in it.

37018

37019 - `type: "slack_workspace"`

37020

37021 - `id: string`

37022

37023 The Bundle Target's identifier. Starts with `btgt_`. No two Bundle Targets have the same one, and the same Bundle and target always give the same one, also after a detach and a new attach. It is opaque.

37024

37025 - `created_at: string`

37026

37027 Time the Bundle was attached.

37028

37029 format: date-time

37030

37031 - `slack_workspace_id: string`

37032

37033 The identifier of the Slack workspace the Bundle is attached to, starting with `slws_`.

37034

37035#### Example

37036

37037```bash

37038curl https://api.anthropic.com/v1/organizations/claude_tag/bundles/$BUNDLE_ID/targets \

37039 -H 'Content-Type: application/json' \

37040 -H 'anthropic-version: 2026-10-21' \

37041 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

37042 -H "Authorization: Bearer $ANTHROPIC_API_KEY" \

37043 -d '{

37044 "slack_channel_id": "xxxxxxxx",

37045 "type": "slack_channel"

37046 }'

37047```

37048

37049##### Response (200)

37050

37051```json

37052{

37053 "id": "id",

37054 "created_at": "2019-12-27T18:11:19.117Z",

37055 "slack_channel_id": "slack_channel_id",

37056 "type": "slack_channel"

37057}

37058```

37059

37060### Get Bundle Target

37061

37062**GET** `/v1/organizations/claude_tag/bundles/{bundle_id}/targets/{target_id}`

37063

37064Get one Bundle Target. Returns 404 when `target_id` is the identifier of another Bundle's Bundle Target, or the Bundle Target was detached.

37065

37066#### Path parameters

37067

37068- `bundle_id: string`

37069

37070 The Bundle's identifier, starting with `capp_`.

37071

37072- `target_id: string`

37073

37074 The Bundle Target's identifier, starting with `btgt_`. Attach Bundle and List Bundle Targets return it. It is never the identifier of the channel or workspace.

37075

37076#### Headers

37077

37078- `"anthropic-version": optional string`

37079

37080 The version of the Claude API you want to use.

37081

37082 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

37083

37084- `"anthropic-beta": optional array of AnthropicBeta`

37085

37086 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

37087

37088 - `string`

37089

37090 - `"message-batches-2024-09-24"`

37091

37092 - `"prompt-caching-2024-07-31"`

37093

37094 - `"computer-use-2024-10-22"`

37095

37096 - `"computer-use-2025-01-24"`

37097

37098 - `"pdfs-2024-09-25"`

37099

37100 - `"token-counting-2024-11-01"`

37101

37102 - `"token-efficient-tools-2025-02-19"`

37103

37104 - `"output-128k-2025-02-19"`

37105

37106 - `"files-api-2025-04-14"`

37107

37108 - `"mcp-client-2025-04-04"`

37109

37110 - `"mcp-client-2025-11-20"`

37111

37112 - `"dev-full-thinking-2025-05-14"`

37113

37114 - `"interleaved-thinking-2025-05-14"`

37115

37116 - `"code-execution-2025-05-22"`

37117

37118 - `"extended-cache-ttl-2025-04-11"`

37119

37120 - `"context-1m-2025-08-07"`

37121

37122 - `"context-management-2025-06-27"`

37123

37124 - `"model-context-window-exceeded-2025-08-26"`

37125

37126 - `"skills-2025-10-02"`

37127

37128 - `"fast-mode-2026-02-01"`

37129

37130 - `"output-300k-2026-03-24"`

37131

37132 - `"user-profiles-2026-03-24"`

37133

37134 - `"user-profiles-2026-08-18"`

37135

37136 - `"user-profiles-2026-09-04"`

37137

37138 - `"advisor-tool-2026-03-01"`

37139

37140 - `"managed-agents-2026-04-01"`

37141

37142 - `"cache-diagnosis-2026-04-07"`

37143

37144 - `"dreaming-2026-04-21"`

37145

37146 - `"thinking-token-count-2026-05-13"`

37147

37148 - `"server-side-fallback-2026-06-01"`

37149

37150 - `"server-side-fallback-2026-07-01"`

37151

37152 - `"fallback-credit-2026-06-01"`

37153

37154 - `"fallback-credit-2026-07-01"`

37155

37156 - `"agent-memory-2026-07-22"`

37157

37158 - `"mid-conversation-tool-changes-2026-07-01"`

37159

37160 - `"compact-2026-01-12"`

37161

37162 - `"computer-use-2025-11-24"`

37163

37164 - `"mcp-tunnels-2026-06-22"`

37165

37166 - `"structured-outputs-2025-11-13"`

37167

37168 - `"task-budgets-2026-03-13"`

37169

37170 - `"thinking-display-updates-2026-08-18"`

37171

37172 - `"ce-user-management-2026-07-13"`

37173

37174 - `"mid-conversation-output-config-2026-07-01"`

37175

37176 - `"thinking-binding-controls-2026-08-01"`

37177

37178 - `"mid-conversation-system-clear-at-2026-08-21"`

37179

37180 - `"compact-2026-09-04"`

37181

37182 - `"inline-tools-2026-09-15"`

37183

37184 - `"mcp-client-2026-09-15"`

37185

37186 - `"ce-plugins-2026-09-01"`

37187

37188 - `"spend-limit-reads-2026-09-26"`

37189

37190 - `"telemetry-destinations-2026-08-11"`

37191

37192 - `"webhooks-2026-10-07"`

37193

37194 - `"message-threads-2026-08-12"`

37195

37196 - `"analytics-usage-cost-2026-08-24"`

37197

37198 - `"experimental-claude-tag-2026-10-21"`

37199

37200#### Returns

37201

37202- `BetaClaudeTagBundleTarget = BetaClaudeTagSlackChannelBundleTarget or BetaClaudeTagSlackWorkspaceBundleTarget`

37203

37204 One Bundle attached to one Slack channel or one Slack workspace. More types of target can be added.

37205

37206 - `BetaClaudeTagSlackChannelBundleTarget object`

37207

37208 One Bundle attached to one Slack channel. The Bundle can be used in that channel only.

37209

37210 - `type: "slack_channel"`

37211

37212 - `id: string`

37213

37214 The Bundle Target's identifier. Starts with `btgt_`. No two Bundle Targets have the same one, and the same Bundle and target always give the same one, also after a detach and a new attach. It is opaque.

37215

37216 - `created_at: string`

37217

37218 Time the Bundle was attached.

37219

37220 format: date-time

37221

37222 - `slack_channel_id: string`

37223

37224 The identifier of the Slack channel the Bundle is attached to, starting with `slch_`. It is the channel object's `id`, not Slack's `channel_id`.

37225

37226 - `BetaClaudeTagSlackWorkspaceBundleTarget object`

37227

37228 One Bundle attached to one Slack workspace. The Bundle can be used in the workspace and in every channel in it.

37229

37230 - `type: "slack_workspace"`

37231

37232 - `id: string`

37233

37234 The Bundle Target's identifier. Starts with `btgt_`. No two Bundle Targets have the same one, and the same Bundle and target always give the same one, also after a detach and a new attach. It is opaque.

37235

37236 - `created_at: string`

37237

37238 Time the Bundle was attached.

37239

37240 format: date-time

37241

37242 - `slack_workspace_id: string`

37243

37244 The identifier of the Slack workspace the Bundle is attached to, starting with `slws_`.

37245

37246#### Example

37247

37248```bash

37249curl https://api.anthropic.com/v1/organizations/claude_tag/bundles/$BUNDLE_ID/targets/$TARGET_ID \

37250 -H 'anthropic-version: 2026-10-21' \

37251 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

37252 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

37253```

37254

37255##### Response (200)

37256

37257```json

37258{

37259 "id": "id",

37260 "created_at": "2019-12-27T18:11:19.117Z",

37261 "slack_channel_id": "slack_channel_id",

37262 "type": "slack_channel"

37263}

37264```

37265

37266### Detach Bundle

37267

37268**DELETE** `/v1/organizations/claude_tag/bundles/{bundle_id}/targets/{target_id}`

37269

37270Detach a Bundle from a Slack channel or a Slack workspace. A repeated detach returns 404.

37271

37272#### Path parameters

37273

37274- `bundle_id: string`

37275

37276 The Bundle's identifier, starting with `capp_`.

37277

37278- `target_id: string`

37279

37280 The Bundle Target's identifier, starting with `btgt_`. Attach Bundle and List Bundle Targets return it. It is never the identifier of the channel or workspace.

37281

37282#### Headers

37283

37284- `"anthropic-version": optional string`

37285

37286 The version of the Claude API you want to use.

37287

37288 Read more about versioning and our version history [here](https://platform.claude.com/docs/en/api/versioning?sdk_languages=http).

37289

37290- `"anthropic-beta": optional array of AnthropicBeta`

37291

37292 This endpoint is in beta: requests must send `experimental-claude-tag-2026-10-21` in this header.

37293

37294 - `string`

37295

37296 - `"message-batches-2024-09-24"`

37297

37298 - `"prompt-caching-2024-07-31"`

37299

37300 - `"computer-use-2024-10-22"`

37301

37302 - `"computer-use-2025-01-24"`

37303

37304 - `"pdfs-2024-09-25"`

37305

37306 - `"token-counting-2024-11-01"`

37307

37308 - `"token-efficient-tools-2025-02-19"`

37309

37310 - `"output-128k-2025-02-19"`

37311

37312 - `"files-api-2025-04-14"`

37313

37314 - `"mcp-client-2025-04-04"`

37315

37316 - `"mcp-client-2025-11-20"`

37317

37318 - `"dev-full-thinking-2025-05-14"`

37319

37320 - `"interleaved-thinking-2025-05-14"`

37321

37322 - `"code-execution-2025-05-22"`

37323

37324 - `"extended-cache-ttl-2025-04-11"`

37325

37326 - `"context-1m-2025-08-07"`

37327

37328 - `"context-management-2025-06-27"`

37329

37330 - `"model-context-window-exceeded-2025-08-26"`

37331

37332 - `"skills-2025-10-02"`

37333

37334 - `"fast-mode-2026-02-01"`

37335

37336 - `"output-300k-2026-03-24"`

37337

37338 - `"user-profiles-2026-03-24"`

37339

37340 - `"user-profiles-2026-08-18"`

37341

37342 - `"user-profiles-2026-09-04"`

37343

37344 - `"advisor-tool-2026-03-01"`

37345

37346 - `"managed-agents-2026-04-01"`

37347

37348 - `"cache-diagnosis-2026-04-07"`

37349

37350 - `"dreaming-2026-04-21"`

37351

37352 - `"thinking-token-count-2026-05-13"`

37353

37354 - `"server-side-fallback-2026-06-01"`

37355

37356 - `"server-side-fallback-2026-07-01"`

37357

37358 - `"fallback-credit-2026-06-01"`

37359

37360 - `"fallback-credit-2026-07-01"`

37361

37362 - `"agent-memory-2026-07-22"`

37363

37364 - `"mid-conversation-tool-changes-2026-07-01"`

37365

37366 - `"compact-2026-01-12"`

37367

37368 - `"computer-use-2025-11-24"`

37369

37370 - `"mcp-tunnels-2026-06-22"`

37371

37372 - `"structured-outputs-2025-11-13"`

37373

37374 - `"task-budgets-2026-03-13"`

37375

37376 - `"thinking-display-updates-2026-08-18"`

37377

37378 - `"ce-user-management-2026-07-13"`

37379

37380 - `"mid-conversation-output-config-2026-07-01"`

37381

37382 - `"thinking-binding-controls-2026-08-01"`

37383

37384 - `"mid-conversation-system-clear-at-2026-08-21"`

37385

37386 - `"compact-2026-09-04"`

37387

37388 - `"inline-tools-2026-09-15"`

37389

37390 - `"mcp-client-2026-09-15"`

37391

37392 - `"ce-plugins-2026-09-01"`

37393

37394 - `"spend-limit-reads-2026-09-26"`

37395

37396 - `"telemetry-destinations-2026-08-11"`

37397

37398 - `"webhooks-2026-10-07"`

37399

37400 - `"message-threads-2026-08-12"`

37401

37402 - `"analytics-usage-cost-2026-08-24"`

37403

37404 - `"experimental-claude-tag-2026-10-21"`

37405

37406#### Returns

37407

37408- `BetaClaudeTagDeletedBundleTarget object`

37409

37410 Confirms that a Bundle was detached.

37411

37412 - `type: "bundle_target_deleted"`

37413

37414 - `id: string`

37415

37416 The identifier the Bundle Target had.

37417

37418#### Example

37419

37420```bash

37421curl https://api.anthropic.com/v1/organizations/claude_tag/bundles/$BUNDLE_ID/targets/$TARGET_ID \

37422 -X DELETE \

37423 -H 'anthropic-version: 2026-10-21' \

37424 -H 'anthropic-beta: experimental-claude-tag-2026-10-21' \

37425 -H "Authorization: Bearer $ANTHROPIC_API_KEY"

37426```

37427

37428##### Response (200)

37429

37430```json

37431{

37432 "id": "id",

37433 "type": "bundle_target_deleted"

37434}

37435```

3295237436

api/beta/organization/analytics Changed · +11 / -7 lines

This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.

Nothing in the body moved in this read. What changed is above.

api/beta/organization/analytics/cost Changed · +4 / -2 lines

from line 429

429429

430430 - `"analytics-usage-cost-2026-08-24"`

431431

432 - `"experimental-claude-tag-2026-10-21"`

433

432434### Returns

433435

434436- `data: array of BetaAnalyticsCostItem`

from line 747

745747 "email_address": "[email protected]",

746748 "name": "Jane Doe",

747749 "type": "user_actor",

748 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

750 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

749751 },

750752 "actor_type": "user_actor",

751753 "agent_id": "agent_01VHd3EAGJ6PLkP9LM7DDu3f",

from line 773

771773 "starting_at": "2026-09-01T00:00:00Z"

772774 }

773775 ],

774 "data_refreshed_at": "2019-12-27T18:11:19.117Z",

776 "data_refreshed_at": "2026-10-01T04:02:11Z",

775777 "has_more": true,

776778 "next_page": "next_page"

777779}

api/beta/organization/analytics/cost/list Changed · +4 / -2 lines

from line 427

427427

428428 - `"analytics-usage-cost-2026-08-24"`

429429

430 - `"experimental-claude-tag-2026-10-21"`

431

430432## Returns

431433

432434- `data: array of BetaAnalyticsCostItem`

from line 745

743745 "email_address": "[email protected]",

744746 "name": "Jane Doe",

745747 "type": "user_actor",

746 "user_id": "user_01WCz1FkmYMm4gnmykNKUu3Q"

748 "user_id": "user_01WxYzAbCdEfGhJkMnPqRsTu"

747749 },

748750 "actor_type": "user_actor",

749751 "agent_id": "agent_01VHd3EAGJ6PLkP9LM7DDu3f",

from line 771

769771 "starting_at": "2026-09-01T00:00:00Z"

770772 }

771773 ],

772 "data_refreshed_at": "2019-12-27T18:11:19.117Z",

774 "data_refreshed_at": "2026-10-01T04:02:11Z",

773775 "has_more": true,

774776 "next_page": "next_page"

775777}