Home Docs Claude Developer Platform

One capture · api

One read of Claude Developer Platformapi-20261011T193717Z

3 pages moved out of 760 read.

Pages moved 3 significant first

Pages read 760 in this capture

Captured 11 Oct 2026 19:37 UTC

Corpus hash 21e8f23e5902 corpus-hash

What this read moved

1-3 of 3

manage-claude/api-and-data-retention Changed · +37 / -37 lines

from line 165

165165* **Yes (qualified):** Your prompts and Claude's outputs are not stored, but a bounded technical artifact (named in the Details column) is retained briefly for the feature to function. See [How Anthropic approaches data retention](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#how-anthropic-approaches-data-retention) for the commitments that govern these features.

166166* **No:** The feature is not eligible. Under HIPAA readiness, the API blocks requests that include a "No" feature and returns a `400` error, unless the feature's Details column says otherwise. Under ZDR, the API does **not** block these features; using one is a choice to step outside your ZDR arrangement for that specific data, and the feature's own documented retention policy applies. Features marked "No" for ZDR are typically stateful (they store jobs, files, or container state), which is why they cannot be zero-retention.

167167

169| -------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------ | ------------------------------------------------------- | ----------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

172| [Advisor tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/advisor-tool) | `/v1/messages` (with `advisor` tool) | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Advisor model output is returned in the API response; nothing is stored server-side after the response. |

173| [Agent skills](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview) | `/v1/messages` (with `skills`) / `/v1/skills` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Skill data retained per standard policy. See [Agent skills](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview#data-retention). |

175| [Batch processing](https://platform.claude.com/docs/en/build-with-claude/batch-processing) | `/v1/messages/batches` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | 29-day retention; async storage required. See [Batch processing](https://platform.claude.com/docs/en/build-with-claude/batch-processing#data-retention). |

176| [Browser use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/browser-use-tool) | `/v1/messages` (with `browser` toolset) | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Client-side tool. Anthropic does not run browser actions or retain page content beyond standard API handling. Not covered under HIPAA readiness; requests that include the browser use tool are not blocked. See [Browser use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/browser-use-tool#data-retention). |

177| [Cache diagnostics](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics) | `/v1/messages` (with `diagnostics`) | <Eligible status="qualified">Yes (qualified)</Eligible> | <Eligible status="no">No</Eligible> | Your prompts and Claude's outputs are not stored. A fingerprint of cryptographic hashes and token-count estimates is retained briefly to enable comparison against the next request. See [Cache diagnostics](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics#data-retention). |

179| [Claude Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) | `/v1/agents`, `/v1/sessions`, `/v1/environments` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including [Self-hosted sandboxes](https://platform.claude.com/docs/en/managed-agents/self-hosted-sandboxes). |

180| [Code execution](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool) | `/v1/messages` (with `code_execution` tool) | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Container data retained up to 30 days. See [Code execution](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool#data-retention). |

181| [Computer use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool) | `/v1/messages` (with `computer` toolset or tool) | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Client-side tool where screenshots and files are captured and stored in your environment, not by Anthropic. See [Computer use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool#data-retention). |

182| [Context editing](https://platform.claude.com/docs/en/build-with-claude/context-editing) | `/v1/messages` (with `context_management`) | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Context edits (tool use clearing and thinking clearing) are applied in real time. |

183| [Context management (compaction)](https://platform.claude.com/docs/en/build-with-claude/compaction-threshold) | `/v1/messages` (with `context_management`) | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Server-side compaction results are returned and round-tripped statelessly through the API response. |

186| [Fast mode](https://platform.claude.com/docs/en/build-with-claude/fast-mode) | `/v1/messages` (with `speed: "fast"`) | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Same Messages API endpoint with faster inference. ZDR applies regardless of speed setting. |

187| [Files API](https://platform.claude.com/docs/en/build-with-claude/files) | `/v1/files` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Files retained until explicitly deleted or they reach their configured expiration. See [Files API](https://platform.claude.com/docs/en/build-with-claude/files#data-retention). |

189| [MCP connector](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector) | `/v1/messages` (with `mcp_servers`) | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Data retained per standard policy. See [MCP connector](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector#data-retention). |

190| [MCP tunnels](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/overview) | `/v1/tunnels` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Research preview. See [MCP tunnels security](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/security) for the data-flow boundary and subprocessor details. |

193| [Mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages) | `/v1/messages` (with `role: "system"` messages) | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Request-shape capability of the Messages API; mid-conversation system messages flow through the standard inference path and nothing is stored server-side after the response. |

195| [Programmatic tool calling](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling) | `/v1/messages` (with `code_execution` tool) | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Built on code execution containers; data retained up to 30 days. See [Programmatic tool calling](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling#data-retention). |

196| [Prompt caching](https://platform.claude.com/docs/en/build-with-claude/prompt-caching) | `/v1/messages` | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Your prompts and Claude's outputs are not stored. KV cache representations and cryptographic hashes are held in memory for the cache TTL and promptly deleted after expiry. See [Prompt caching](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#data-retention). |

198| [Structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs) | `/v1/messages` | <Eligible status="qualified">Yes (qualified)</Eligible> | <Eligible>Yes</Eligible> | Your prompts and Claude's outputs are not stored. Only the JSON schema is cached, for up to 24 hours since last use. This also covers [strict tool use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/strict-tool-use) (`strict: true` on tools), which uses the same grammar pipeline. PHI must not be included in JSON schema definitions; see [PHI handling guidelines](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#phi-handling-guidelines). See [Structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs#data-retention). |

202| [Tool search](https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-search-tool) | `/v1/messages` (with `tool_search` tool) | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Server-side tool executed by Anthropic; the tool definitions in the request are searched in memory per call and nothing is stored after the response. |

203| [Web fetch](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool) | `/v1/messages` (with `web_fetch` tool) | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Fetched web content returned in the API response. [Dynamic filtering](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool#dynamic-filtering) is not eligible for ZDR or HIPAA. Website publishers may retain request data (such as fetched URLs and request metadata) according to their own policies. |

204| [Web search](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool) | `/v1/messages` (with `web_search` tool) | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Real-time web search results returned in the API response. [Dynamic filtering](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool#dynamic-filtering) is not eligible for ZDR or HIPAA. |

169| -------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------- | ------------------------------------------------------- | ----------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

172| [Advisor tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/advisor-tool) | `/v1/messages` with `advisor` tool | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Advisor model output is returned in the API response; nothing is stored server-side after the response. |

173| [Agent skills](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview) | `/v1/skills` `/v1/messages` with `skills` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Skill data retained per standard policy. See [Agent skills](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview#data-retention). |

176| [Browser use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/browser-use-tool) | `/v1/messages` with `browser` toolset | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Client-side tool. Anthropic does not run browser actions or retain page content beyond standard API handling. Not covered under HIPAA readiness; requests that include the browser use tool are not blocked. See [Browser use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/browser-use-tool#data-retention). |

177| [Cache diagnostics](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics) | `/v1/messages` with `diagnostics` | <Eligible status="qualified">Yes (qualified)</Eligible> | <Eligible status="no">No</Eligible> | Your prompts and Claude's outputs are not stored. A fingerprint of cryptographic hashes and token-count estimates is retained briefly to enable comparison against the next request. See [Cache diagnostics](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics#data-retention). |

179| [Claude Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) | `/v1/agents` `/v1/sessions` `/v1/environments` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including [Self-hosted sandboxes](https://platform.claude.com/docs/en/managed-agents/self-hosted-sandboxes). |

180| [Code execution](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool) | `/v1/messages` with `code_execution` tool | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Container data retained up to 30 days. See [Code execution](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool#data-retention). |

181| [Computer use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool) | `/v1/messages` with `computer` toolset or tool | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Client-side tool where screenshots and files are captured and stored in your environment, not by Anthropic. See [Computer use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool#data-retention). |

182| [Context editing](https://platform.claude.com/docs/en/build-with-claude/context-editing) | `/v1/messages` with `context_management` | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Context edits (tool use clearing and thinking clearing) are applied in real time. |

183| [Context management (compaction)](https://platform.claude.com/docs/en/build-with-claude/compaction-threshold) | `/v1/messages` with `context_management` | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Server-side compaction results are returned and round-tripped statelessly through the API response. |

186| [Fast mode](https://platform.claude.com/docs/en/build-with-claude/fast-mode) | `/v1/messages` with `speed: "fast"` | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Same Messages API endpoint with faster inference. ZDR applies regardless of speed setting. |

189| [MCP connector](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector) | `/v1/messages` with `mcp_servers` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Data retained per standard policy. See [MCP connector](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector#data-retention). |

193| [Mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages) | `/v1/messages` with `role: "system"` messages | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Request-shape capability of the Messages API; mid-conversation system messages flow through the standard inference path and nothing is stored server-side after the response. |

195| [Programmatic tool calling](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling) | `/v1/messages` with `code_execution` tool | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Built on code execution containers; data retained up to 30 days. See [Programmatic tool calling](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling#data-retention). |

202| [Tool search](https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-search-tool) | `/v1/messages` with `tool_search` tool | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Server-side tool executed by Anthropic; the tool definitions in the request are searched in memory per call and nothing is stored after the response. |

203| [Web fetch](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool) | `/v1/messages` with `web_fetch` tool | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Fetched web content returned in the API response. [Dynamic filtering](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool#dynamic-filtering) is not eligible for ZDR or HIPAA. Website publishers may retain request data (such as fetched URLs and request metadata) according to their own policies. |

204| [Web search](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool) | `/v1/messages` with `web_search` tool | <Eligible>Yes</Eligible> | <Eligible>Yes</Eligible> | Real-time web search results returned in the API response. [Dynamic filtering](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool#dynamic-filtering) is not eligible for ZDR or HIPAA. |

205205

206206## Retention regardless of arrangement

207207

managed-agents/scheduled-deployments Changed · +131 / -15 lines

from line 272

272272

273273The following example sets a budget on an existing deployment:

274274

275```bash cURL

276curl --fail-with-body -sS "https://api.anthropic.com/v1/deployments/$DEPLOYMENT_ID?beta=true" \

277 -H "x-api-key: $ANTHROPIC_API_KEY" \

278 -H "anthropic-version: 2023-06-01" \

279 -H "anthropic-beta: managed-agents-2026-04-01" \

280 -H "content-type: application/json" \

281 -d @- <<'EOF'

282{

283 "budget": {

284 "type": "limit",

285 "max_list_cost": {"amount": "2000", "currency": "USD"}

275<CodeGroup defaultLanguage="CLI">

276 ```bash cURL

277 curl --fail-with-body -sS "https://api.anthropic.com/v1/deployments/$DEPLOYMENT_ID?beta=true" \

278 -H "x-api-key: $ANTHROPIC_API_KEY" \

279 -H "anthropic-version: 2023-06-01" \

280 -H "anthropic-beta: managed-agents-2026-04-01" \

281 -H "content-type: application/json" \

282 -d @- <<'EOF'

283 {

284 "budget": {

285 "type": "limit",

286 "max_list_cost": {"amount": "2000", "currency": "USD"}

287 }

286288 }

287}

288EOF

289```

289 EOF

290 ```

291

292 <CodeGroupItem>

293 ```bash CLI

294 ant apply deployment.md

295 ```

296

297 <File filename="deployment.md">

298 ```markdown

299 ---

300 name: Weekly compliance scan

301 agent: agent_011CYm1BLqPXpQRk5khsSXrs

302 environment_id: env_01595EKxaaTTGwwY3kyXdtbs

303 schedule:

304 type: cron

305 expression: "0 20 * * 5"

306 timezone: America/New_York

307 budget:

308 type: limit

309 max_list_cost:

310 amount: "2000"

311 currency: USD

312 ---

313

314 Run the weekly compliance scan.

315 ```

316 </File>

317 </CodeGroupItem>

318

319 ```python Python

320 updated_deployment = client.beta.deployments.update(

321 deployment.id,

322 budget={

323 "type": "limit",

324 "max_list_cost": {"amount": "2000", "currency": "USD"},

325 },

326 )

327 print(updated_deployment.budget.max_list_cost.amount) # 2000

328 ```

329

330 ```typescript TypeScript

331 const updatedDeployment = await client.beta.deployments.update(deployment.id, {

332 budget: {

333 type: "limit",

334 max_list_cost: { amount: "2000", currency: "USD" },

335 },

336 });

337 console.log(updatedDeployment.budget?.max_list_cost.amount); // 2000

338 ```

339

340 ```csharp C#

341 var updatedDeployment = await client.Beta.Deployments.Update(deployment.ID, new()

342 {

343 Budget = new()

344 {

345 Type = BetaManagedAgentsBudgetLimitType.Limit,

346 MaxListCost = new() { Amount = "2000", Currency = BetaCurrency.Usd },

347 },

348 });

349 Console.WriteLine(updatedDeployment.Budget?.MaxListCost.Amount); // 2000

350 ```

351

352 ```go Go

353 updatedDeployment, err := client.Beta.Deployments.Update(ctx, deployment.ID, anthropic.BetaDeploymentUpdateParams{

354 Budget: anthropic.BetaManagedAgentsBudgetLimitParam{

355 Type: anthropic.BetaManagedAgentsBudgetLimitTypeLimit,

356 MaxListCost: anthropic.BetaMonetaryAmountParam{

357 Amount: "2000",

358 Currency: anthropic.BetaCurrencyUsd,

359 },

360 },

361 })

362 if err != nil {

363 panic(err)

364 }

365 fmt.Println(updatedDeployment.Budget.MaxListCost.Amount) // 2000

366 ```

367

368 ```java Java

369 var updatedDeployment = client.beta().deployments().update(

370 deployment.id(),

371 DeploymentUpdateParams.builder()

372 .budget(BetaManagedAgentsBudgetLimit.builder()

373 .type(BetaManagedAgentsBudgetLimit.Type.LIMIT)

374 .maxListCost(BetaMonetaryAmount.builder()

375 .amount("2000")

376 .currency(BetaCurrency.USD)

377 .build())

378 .build())

379 .build()

380 );

381 IO.println(updatedDeployment.budget().orElseThrow().maxListCost().amount()); // 2000

382 ```

383

384 ```php PHP

385 $updatedDeployment = $client->beta->deployments->update(

386 $deployment->id,

387 budget: [

388 'type' => 'limit',

389 'max_list_cost' => ['amount' => '2000', 'currency' => 'USD'],

390 ],

391 );

392 echo "{$updatedDeployment->budget->maxListCost->amount}\n"; // 2000

393 ```

394

395 ```ruby Ruby

396 updated_deployment = client.beta.deployments.update(

397 deployment.id,

398 budget: {

399 type: "limit",

400 max_list_cost: {amount: "2000", currency: "USD"}

401 }

402 )

403 puts updated_deployment.budget.max_list_cost.amount # 2000

404 ```

405</CodeGroup>

290406

291407## Deployment runs

292408

managed-agents/sessions Changed · +128 / -21 lines

from line 800

800800

801801### Set a session budget

802802

803To cap what a session can spend, pass the optional `budget` object when you create it. A budget is a hard ceiling on the session's list cost: the platform prices everything the session consumes at public list rates, and the session stops issuing new model requests once that running total reaches `max_list_cost`. Set `type` to `limit` and give `max_list_cost` an `amount` and a `currency`. `amount` is a whole number of US cents written as a string, such as `"2500"` for $25.00; the API takes a string rather than a number so no floating-point rounding is ever applied. `USD` is the only currency currently supported. When the session reaches the cap, it pauses and goes idle with the stop reason `budget_reached`. The cap is enforced between model requests, so the request that crosses it finishes first and the session's final list cost can land [a fraction past the cap](https://platform.claude.com/docs/en/managed-agents/budgets#when-a-session-reaches-its-budget). A budget can only be attached at creation: you can [change or remove](https://platform.claude.com/docs/en/managed-agents/session-operations#updating-the-session-budget) it later, but you can't add one to a session created without it.

803To cap what a session can spend, pass the optional `budget` object when you create it. A budget is a hard ceiling on the session's list cost: the platform prices everything the session consumes at public list rates, and the session stops issuing new model requests once that running total reaches `max_list_cost`. Set `type` to `limit` and give `max_list_cost` an `amount` and a `currency`. `amount` is a whole number of US cents written as a string, such as `"125"` for $1.25. The API takes a string rather than a number so no floating-point rounding is ever applied. `USD` is the only currency currently supported. When the session reaches the cap, it pauses and goes idle with the stop reason `budget_reached`. The cap is enforced between model requests, so the request that crosses it finishes first and the session's final list cost can land [a fraction past the cap](https://platform.claude.com/docs/en/managed-agents/budgets#when-a-session-reaches-its-budget). A budget can only be attached at creation: you can [change or remove](https://platform.claude.com/docs/en/managed-agents/session-operations#updating-the-session-budget) it later, but you can't add one to a session created without it.

804804

805The following example creates a session with a $25.00 budget; the response echoes the `budget` on the session resource:

805The following example creates a session with a $1.25 budget:

806806

807```bash cURL

808curl -fsSL https://api.anthropic.com/v1/sessions \

809 -H "x-api-key: $ANTHROPIC_API_KEY" \

810 -H "anthropic-version: 2023-06-01" \

811 -H "anthropic-beta: managed-agents-2026-04-01" \

812 -H "content-type: application/json" \

813 -d @- <<EOF

814{

815 "agent": "$AGENT_ID",

816 "environment_id": "$ENVIRONMENT_ID",

817 "budget": {

818 "type": "limit",

819 "max_list_cost": {"amount": "2500", "currency": "USD"}

807<CodeGroup>

808 ```bash cURL

809 curl -sS --fail-with-body https://api.anthropic.com/v1/sessions \

810 -H "x-api-key: $ANTHROPIC_API_KEY" \

811 -H "anthropic-version: 2023-06-01" \

812 -H "anthropic-beta: managed-agents-2026-04-01" \

813 -H "content-type: application/json" \

814 -d @- <<EOF

815 {

816 "agent": "$AGENT_ID",

817 "environment_id": "$ENVIRONMENT_ID",

818 "budget": {

819 "type": "limit",

820 "max_list_cost": {"amount": "125", "currency": "USD"}

821 }

820822 }

821}

822EOF

823```

823 EOF

824 ```

825

826 ```bash CLI

827 # Keep the amount quoted so it is sent as a string, not a number.

828 ant beta:sessions create \

829 --agent "$AGENT_ID" \

830 --environment-id "$ENVIRONMENT_ID" \

831 --budget '{type: limit, max_list_cost: {amount: "125", currency: USD}}'

832 ```

833

834 ```python Python

835 session = client.beta.sessions.create(

836 agent=agent.id,

837 environment_id=environment.id,

838 budget={

839 "type": "limit",

840 "max_list_cost": {"amount": "125", "currency": "USD"},

841 },

842 )

843 print(session.id, session.budget.max_list_cost.amount) # sesn_01... 125

844 ```

845

846 ```typescript TypeScript

847 const session = await client.beta.sessions.create({

848 agent: agent.id,

849 environment_id: environment.id,

850 budget: {

851 type: "limit",

852 max_list_cost: { amount: "125", currency: "USD" }

853 }

854 });

855 console.log(session.id, session.budget?.max_list_cost.amount); // sesn_01... 125

856 ```

857

858 ```csharp C#

859 var session = await client.Beta.Sessions.Create(new()

860 {

861 Agent = agent.ID,

862 EnvironmentID = environment.ID,

863 Budget = new()

864 {

865 Type = BetaManagedAgentsBudgetLimitType.Limit,

866 MaxListCost = new() { Amount = "125", Currency = BetaCurrency.Usd },

867 },

868 });

869 Console.WriteLine($"{session.ID} {session.Budget?.MaxListCost.Amount}"); // sesn_01... 125

870 ```

871

872 ```go Go

873 session, err := client.Beta.Sessions.New(ctx, anthropic.BetaSessionNewParams{

874 Agent: anthropic.BetaSessionNewParamsAgentUnion{

875 OfString: anthropic.String(agent.ID),

876 },

877 EnvironmentID: environment.ID,

878 Budget: anthropic.BetaManagedAgentsBudgetLimitParam{

879 Type: anthropic.BetaManagedAgentsBudgetLimitTypeLimit,

880 MaxListCost: anthropic.BetaMonetaryAmountParam{

881 Amount: "125",

882 Currency: anthropic.BetaCurrencyUsd,

883 },

884 },

885 })

886 if err != nil {

887 panic(err)

888 }

889 fmt.Println(session.ID, session.Budget.MaxListCost.Amount) // sesn_01... 125

890 ```

891

892 ```java Java

893 var session = client.beta().sessions().create(SessionCreateParams.builder()

894 .agent(agent.id())

895 .environmentId(environment.id())

896 .budget(BetaManagedAgentsBudgetLimit.builder()

897 .type(BetaManagedAgentsBudgetLimit.Type.LIMIT)

898 .maxListCost(BetaMonetaryAmount.builder()

899 .amount("125")

900 .currency(BetaCurrency.USD)

901 .build())

902 .build())

903 .build());

904 IO.println(session.id() + " " + session.budget().orElseThrow().maxListCost().amount()); // sesn_01... 125

905 ```

906

907 ```php PHP

908 $session = $client->beta->sessions->create(

909 agent: $agent->id,

910 environmentID: $environment->id,

911 budget: [

912 'type' => 'limit',

913 'max_list_cost' => ['amount' => '125', 'currency' => 'USD'],

914 ],

915 );

916 echo "{$session->id} {$session->budget->maxListCost->amount}\n"; // sesn_01... 125

917 ```

918

919 ```ruby Ruby

920 session = client.beta.sessions.create(

921 agent: agent.id,

922 environment_id: environment.id,

923 budget: {

924 type: "limit",

925 max_list_cost: {amount: "125", currency: "USD"}

926 }

927 )

928 puts "#{session.id} #{session.budget.max_list_cost.amount}" # sesn_01... 125

929 ```

930</CodeGroup>

824931

825932See [Session budgets](https://platform.claude.com/docs/en/managed-agents/budgets) for how enforcement works, what counts toward list cost, and how budgets behave in multiagent sessions.

826933