One read of Claude Developer Platformapi-20261029T173726Z

This capture is too large to show at once. Changes 301-320 of 320 are below, significant first; the rest are on the following screens.

build-with-claude/fallback-credit Changed · +1 / -1 lines

from line 41

4141 On a refusal, `stop_details` includes two fields:

4242 

4343 * **`fallback_credit_token`:** an opaque string that represents the credit

44 * **`fallback_has_prefill_claim`:** a Boolean that tells you which retry body shape to use

44 * **`fallback_has_prefill_claim`:** a boolean that tells you which retry body shape to use

4545 

4646 Both are `null` when no credit is available for the refusal.

4747 </Step>

build-with-claude/fast-mode Changed · +1 / -1 lines

from line 356

356356}

357357```

358358 

359To track fast mode usage and costs across your organization, see the [Usage and Cost API](https://platform.claude.com/docs/en/manage-claude/usage-cost-api).

359To track fast mode usage and costs across your organization, see the [Usage and Cost Analytics API](https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api).

360360 

361361## Retries and fallbacks

362362 

build-with-claude/overview Changed · +1 / -1 lines

from line 6

66 

77For each feature’s eligibility under zero data retention (ZDR) and HIPAA readiness, see [API and data retention](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility).

88 

9For administration and governance, see the [Admin API](https://platform.claude.com/docs/en/manage-claude/admin-api), the [Usage and Cost API](https://platform.claude.com/docs/en/manage-claude/usage-cost-api), and the [Compliance API](https://platform.claude.com/docs/en/manage-claude/compliance-api).

9For administration and governance, see the [Admin API](https://platform.claude.com/docs/en/manage-claude/admin-api), the [Usage and Cost Analytics API](https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api), and the [Compliance API](https://platform.claude.com/docs/en/manage-claude/compliance-api).

1010 

1111## Platform features

1212 

build-with-claude/prompt-engineering/claude-prompting-best-practices Changed · +1 / -1 lines

from line 1127

11271127 

112811286. **Tune anti-laziness prompting:** If your prompts previously encouraged the model to be more thorough or use tools more aggressively, reduce that guidance. Claude 4.6 models are more proactive and might overtrigger on instructions that were needed for previous models.

11291129 

11307. **Pass thinking blocks back unchanged and keep history append-only:** Append each assistant turn exactly as the API returned it, thinking blocks included. On Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5, and Claude Haiku 5.5, [modifying the conversation before a thinking block](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-in-conversation) results in an error, or in the block being dropped if you opt into that: editing earlier messages, rebuilding `system` or `tools`, or summarizing older turns in place between requests invalidates every later thinking block, so move those changes to mid-conversation system messages and server-side context management. See [Keep the conversation history append-only](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-fable-5-1#keep-the-conversation-history-append-only).

11307. **Pass thinking blocks back unchanged and keep history append-only:** Append each assistant turn exactly as the API returned it, thinking blocks included. On Claude Fable 5.1, Claude Mythos 5.1, Claude Opus 5.5, Claude Sonnet 5.5, and Claude Haiku 5.5, editing earlier messages, rebuilding `system` or `tools`, or summarizing older turns in place between requests [invalidates every later thinking block](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-in-conversation). Move those changes to mid-conversation system messages and server-side context management. See [Keep the conversation history append-only](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-fable-5-1#keep-the-conversation-history-append-only).

11311131 

11321132For detailed migration steps, see the [Migration guide](https://platform.claude.com/docs/en/about-claude/models/migration-guide).

11331133 

build-with-claude/structured-outputs Changed · +1 / -1 lines

from line 2623

26232623 

26242624<Accordion title="Supported features">

26252625 * All basic types: object, array, string, integer, number, boolean, null

2626 * `enum` (strings, numbers, Booleans, or nulls only, no complex types). For a capitalization caveat, see [Invalid outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs#invalid-outputs).

2626 * `enum` (strings, numbers, booleans, or nulls only, no complex types). For a capitalization caveat, see [Invalid outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs#invalid-outputs).

26272627 * `const`

26282628 * `anyOf` and `allOf` (with limitations: `allOf` with `$ref` not supported)

26292629 * `$ref`, `$def`, and `definitions` (external `$ref` not supported)

build-with-claude/vision Changed · +1 / -1 lines

from line 1274

12741274 

12751275The maximum dimensions for each image are 8000x8000 px.

12761276 

1277If a single API request contains more than 20 images, a stricter per-image dimension limit applies to every image in that request. All `image` blocks in the request count toward this threshold, including images from earlier conversation turns that you resend and images nested inside `tool_result` content (for example, screenshots returned to the computer use tool). On Amazon Bedrock and Claude Models on Google Cloud (Google managed), document blocks such as PDFs also count toward this threshold. Images exceeding the stricter limit are rejected with an `invalid_request_error` whose message references “many-image requests” and states the current limit in pixels. To stay under the limit on all platforms, either resize each image so that neither dimension exceeds 2000 px, or keep the request to 20 or fewer image and document blocks.

1277If a single API request contains more than 20 images, a stricter per-image dimension limit applies to every image in that request. All `image` blocks in the request count toward this threshold, including images from earlier conversation turns that you resend and images nested inside `tool_result` content (for example, screenshots returned to the computer use tool). On Amazon Bedrock and Claude Models on Google Cloud (Google managed), document blocks such as PDFs also count toward this threshold. Images exceeding the stricter limit are rejected with an `invalid_request_error` whose message references “many-image requests” and states the current limit in pixels. To stay under the limit on all platforms, either resize each image so that neither dimension exceeds 3,000 px, or keep the request to 20 or fewer image and document blocks.

12781278 

12791279The maximum size per image is:

12801280 

home Changed · +2 / -2 lines

from line 339

339339 API keys in Claude Console

340340 </HomeJourneyLink>

341341 

342 <HomeJourneyLink icon="chart" href="https://platform.claude.com/docs/en/manage-claude/usage-cost-api">

342 <HomeJourneyLink icon="chart" href="https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api">

343343 Usage monitoring

344344 </HomeJourneyLink>

345345 

from line 397

397397 API keys in Claude Console

398398 </HomeJourneyLink>

399399 

400 <HomeJourneyLink icon="chart" href="https://platform.claude.com/docs/en/manage-claude/usage-cost-api">

400 <HomeJourneyLink icon="chart" href="https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api">

401401 Usage monitoring

402402 </HomeJourneyLink>

403403 </HomeJourneyStep>

manage-claude/claude-code-analytics-api Changed · +1 / -1 lines

from line 283

283283* [Admin API](https://platform.claude.com/docs/en/manage-claude/admin-api)

284284* [Admin API reference](https://platform.claude.com/docs/en/api/beta/organization)

285285* [Claude Code Analytics dashboard](https://platform.claude.com/claude-code) in the Claude Console

286* [Usage and Cost API](https://platform.claude.com/docs/en/manage-claude/usage-cost-api) - Track API usage across all Anthropic services

286* [Usage and Cost Analytics API](https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api) - Track token usage and cost across your organization

287287* [Compliance API](https://platform.claude.com/docs/en/manage-claude/compliance-api) - Retrieve audit and activity data

288288* [Identity and access management](https://code.claude.com/docs/en/iam)

289289* [Monitoring usage with OpenTelemetry](https://code.claude.com/docs/en/monitoring-usage) for custom metrics and alerting

manage-claude/claude-tag-api Changed · +1 / -1 lines

from line 259

259259Settings exist at three levels: the channel, the workspace the channel is in, and your organization’s Slack defaults, which an organization owner sets in claude.ai. Each level stores only what was set on it. The `settings` of a channel or workspace object can hold either of two things:

260260 

261261* **What it stores.** This is what a response holds unless you ask otherwise. A setting that is not set on the channel or workspace is `null`. Use it to compare against configuration you keep elsewhere.

262* **What applies in it**, once every level is taken into account. To get this, add the query parameter `include_inherited=true` to any `GET` request that returns a channel or a workspace, the two lists included. Use it to audit, or to find out why Claude behaves as it does in one channel. The parameter defaults to `false`, and a value that is not a Boolean returns `400`. A request that sets a channel up or changes settings does not take the parameter and returns `400` if you send it; its response holds what is stored.

262* **What applies in it**, once every level is taken into account. To get this, add the query parameter `include_inherited=true` to any `GET` request that returns a channel or a workspace, the two lists included. Use it to audit, or to find out why Claude behaves as it does in one channel. The parameter defaults to `false`, and a value that is not a boolean returns `400`. A request that sets a channel up or changes settings does not take the parameter and returns `400` if you send it; its response holds what is stored.

263263 

264264In both cases `sources` names, for each setting, the level that the value in `settings` comes from. Without `include_inherited`, the source of a setting that the channel or workspace stores names the channel or workspace itself, and a setting it does not store has no source: `null`, or an empty list for `instructions`. With `include_inherited=true`, every setting except `instructions` has a source, and a setting that no level sets has the source `{"type": "default"}`; when no level sets `instructions`, it is `null` and its source list is empty. So to tell a value the channel or workspace sets itself from one it inherits, look at its source, not at whether the value is `null`.

265265 

manage-claude/data-residency Changed · +2 / -2 lines

from line 352

352352 Learn about workspace configuration.

353353 </Card>

354354 

355 <Card title="Usage and Cost API" icon="chart" href="https://platform.claude.com/docs/en/manage-claude/usage-cost-api">

356 Track usage and costs by data residency.

355 <Card title="Usage and Cost Analytics API" icon="chart" href="https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api">

356 Track usage and costs by inference geo.

357357 </Card>

358358</CardGroup>

359359 

manage-claude/opentelemetry Changed · +1 / -1 lines

from line 261

261261`gen_ai.usage.input_tokens` already includes the tokens read from and written to the prompt cache, so don’t add the cache attributes to it. For all usage attributes, see [Token usage](https://platform.claude.com/docs/en/api/opentelemetry/spans#token-usage).

262262 

263263<Note>

264 Exported spans aren’t guaranteed to be a complete record of your API usage. Their delivery is [best effort](https://platform.claude.com/docs/en/manage-claude/opentelemetry/api#delivery-and-health), so a span can be lost. A caller can also leave out a request’s spans by turning off the sampled flag in the request’s [`traceparent` header](https://platform.claude.com/docs/en/manage-claude/opentelemetry/api#connect-api-spans-to-your-own-traces). To track each API key’s usage, for example to spot a leaked key, use the [Usage API](https://platform.claude.com/docs/en/manage-claude/usage-cost-api#usage-api), which also counts the usage of requests without spans.

264 Exported spans aren’t guaranteed to be a complete record of your API usage. Their delivery is [best effort](https://platform.claude.com/docs/en/manage-claude/opentelemetry/api#delivery-and-health), so a span can be lost. A caller can also leave out a request’s spans by turning off the sampled flag in the request’s [`traceparent` header](https://platform.claude.com/docs/en/manage-claude/opentelemetry/api#connect-api-spans-to-your-own-traces). To track each API key’s usage, for example to spot a leaked key, use the [Usage and Cost Analytics API](https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api) and group by `api_key_id`. It also counts the usage of requests without spans. Usage typically appears there within 4 hours (see [Data freshness](https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api#data-freshness)).

265265</Note>

266266 

267267## Availability

manage-claude/user-profiles Changed · +2 / -2 lines

from line 688

688688 Choose how your services authenticate to the Claude API.

689689 </Card>

690690 

691 <Card title="Usage and Cost API" href="https://platform.claude.com/docs/en/manage-claude/usage-cost-api">

692 Track API token usage and costs for your organization.

691 <Card title="Usage and Cost Analytics API" href="https://platform.claude.com/docs/en/manage-claude/usage-cost-analytics-api">

692 Track token usage and costs for your organization.

693693 </Card>

694694 

695695 <Card title="Data residency" href="https://platform.claude.com/docs/en/manage-claude/data-residency">

manage-claude/wif-reference Changed · +1 / -1 lines

from line 192

192192| ---------------- | -------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |

193193| `subject_prefix` | string | Exact match against the JWT `sub` claim. A trailing `*` makes it a prefix match (the `sub` value must start with the characters before the `*`). Case-sensitive. |

194194| `audience` | string | The JWT `aud` claim must contain this exact string. When `aud` is an array, any element matching exactly satisfies the check. |

195| `claims` | map\<string, string> | Each key is a top-level claim name and each value is the required exact string value. For nested, numeric, Boolean, or complex claims such as lists and maps, use `condition` with a CEL expression instead. |

195| `claims` | map\<string, string> | Each key is a top-level claim name and each value is the required exact string value. For nested, numeric, boolean, or complex claims such as lists and maps, use `condition` with a CEL expression instead. |

196196| `condition` | string (CEL) | A [CEL](https://cel.dev/) expression that must evaluate to `true`. |

197197 

198198### CEL evaluation environment

managed-agents/vaults Changed · +1 / -1 lines

from line 705

705705 

706706 Request payloads are often assembled from content the agent is working with, so the request body is the broader exposure surface. Most services read an API key from a request header, so enabling only `header` (csharp, go: `Header`) is the narrower configuration. It scopes substitution to request header values for that credential.

707707 

708 The credential’s `injection_location` (java, php: `injectionLocation`; csharp, go: `InjectionLocation`) controls which parts of an outbound request the secret is substituted into. It is an optional object, a sibling of `networking` (csharp, go: `Networking`), with two Boolean fields: `header` (csharp, go: `Header`) (request headers) and `body` (csharp, go: `Body`) (request body). `injection_location` (java, php: `injectionLocation`; csharp, go: `InjectionLocation`) is independent of `allowed_hosts` (java, php: `allowedHosts`; csharp, go: `AllowedHosts`) in `networking` (csharp, go: `Networking`): `allowed_hosts` (java, php: `allowedHosts`; csharp, go: `AllowedHosts`) scopes which hosts the secret is substituted for, and `injection_location` (java, php: `injectionLocation`; csharp, go: `InjectionLocation`) scopes which parts of the request it is substituted into.

708 The credential’s `injection_location` (java, php: `injectionLocation`; csharp, go: `InjectionLocation`) controls which parts of an outbound request the secret is substituted into. It is an optional object, a sibling of `networking` (csharp, go: `Networking`), with two boolean fields: `header` (csharp, go: `Header`) (request headers) and `body` (csharp, go: `Body`) (request body). `injection_location` (java, php: `injectionLocation`; csharp, go: `InjectionLocation`) is independent of `allowed_hosts` (java, php: `allowedHosts`; csharp, go: `AllowedHosts`) in `networking` (csharp, go: `Networking`): `allowed_hosts` (java, php: `allowedHosts`; csharp, go: `AllowedHosts`) scopes which hosts the secret is substituted for, and `injection_location` (java, php: `injectionLocation`; csharp, go: `InjectionLocation`) scopes which parts of the request it is substituted into.

709709 

710710 `injection_location` (java, php: `injectionLocation`; csharp, go: `InjectionLocation`) behaves differently on create and on update:

711711 

models/haiku-5-5/migration-guide Changed · +2 / -2 lines

from line 133

133133 

134134## Replay thinking blocks through the account that produced them

135135 

136Thinking blocks from Claude Haiku 5.5 work only in the account that produced them, or in an account linked to it. When another account sends one of these blocks, the API drops the block before the model receives it, and the request succeeds without that reasoning. This affects code that stores conversations and replays them through a different account, for example a service that serves several customers from one conversation store. Replay each conversation through the account that produced it. See [Thinking blocks stay with the account that produced them](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#account-bound-thinking).

136Thinking blocks from Claude Haiku 5.5 work only in the account that produced them or in an account linked to it. When another account sends one of these blocks, the API drops the block before the model receives it, and the request succeeds without that reasoning. This affects code that stores conversations and replays them through a different account, for example a service that serves several customers from one conversation store. Replay each conversation through the account that produced it. See [Thinking blocks stay with the account that produced them](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#account-bound-thinking).

137137 

138138## Keep earlier turns unchanged

139139 

140A Claude Haiku 5.5 thinking block stays valid only while everything sent before it is unchanged: a request that sends a thinking block back after a change to `system`, `tools`, or earlier `messages` returns a 400 error. Claude Haiku 4.5 doesn’t run this check. Keep conversations append-only. On accounts created before August 31, 2026, 00:00 UTC, the error comes only on requests that set `thinking.block_binding.prefix_mismatch_behavior`. For the changes that trigger the error and what to do instead, see [Who needs to change anything](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#who-is-affected).

140A Claude Haiku 5.5 thinking block stays valid only while everything sent before it is unchanged: the `system` prompt, the `tools`, and every earlier message. Claude Haiku 4.5 doesn’t run the prefix check. Keep conversations append-only. A request that sends a block back after such a change and leaves `thinking.block_binding.prefix_mismatch_behavior` unset gets a 400 error on accounts created on or after August 31, 2026, 00:00 UTC. On accounts created before that date, the request gets the error only when it sets the field to `"error"`. For the changes that trigger the error and what to do instead, see [Who the prefix check affects](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#who-is-affected).

141141 

142142## Migrating to Claude Haiku 5.5 from Claude Haiku 3.5 and earlier Haiku models

143143 

models/opus-5-5/migration-guide Changed · +1 / -1 lines

from line 876

876876 

877877#### Thinking blocks are tied to the model and the conversation

878878 

879On the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read Claude Opus 5.5 thinking blocks; no other model does. A router or fallback that moves a conversation from Claude Opus 5.5 to any other model runs those turns without them. In the other direction, Claude Opus 5.5 reads thinking blocks from Claude Opus 5, from earlier Opus, Sonnet, and Haiku models, and, on the Claude API and Claude Models on Google Cloud (Google managed), from Claude Sonnet 5.5 and Claude Haiku 5.5, but not from Claude Fable or Claude Mythos models. Keep the conversation append-only (no edits to the `system` prompt, `tools`, or earlier messages mid-conversation) so the blocks stay valid; Claude Code, claude.ai, Claude Managed Agents, and the Claude Agent SDK already do. Enforcement matches Claude Fable 5.1 on every platform: for accounts created on or after August 31, 2026, 00:00 UTC, replaying a thinking block after such an edit returns a 400 error by default. There is no code change for append-only integrations. See [Thinking blocks are tied to the model and the conversation](https://platform.claude.com/docs/en/models/opus-5-5/whats-new-opus-5-5#thinking-blocks-are-tied-to-the-model-that-produced-them) and [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking).

879On the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read Claude Opus 5.5 thinking blocks; no other model does. A router or fallback that moves a conversation from Claude Opus 5.5 to any other model runs those turns without them. In the other direction, Claude Opus 5.5 reads thinking blocks from Claude Opus 5, from earlier Opus, Sonnet, and Haiku models, and, on the Claude API and Claude Models on Google Cloud (Google managed), from Claude Sonnet 5.5 and Claude Haiku 5.5, but not from Claude Fable or Claude Mythos models. Keep the conversation append-only (no edits to the `system` prompt, `tools`, or earlier messages mid-conversation) so the blocks stay valid; Claude Code, claude.ai, Claude Managed Agents, and the Claude Agent SDK already do. On accounts created on or after August 31, 2026, 00:00 UTC, on the Claude API and on cloud platforms, a request that replays a thinking block after such an edit and leaves `thinking.block_binding.prefix_mismatch_behavior` unset gets a 400 error. There is no code change for append-only integrations. See [Thinking blocks are tied to the model and the conversation](https://platform.claude.com/docs/en/models/opus-5-5/whats-new-opus-5-5#thinking-blocks-are-tied-to-the-model-that-produced-them) and [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking).

880880 

881881#### The `computer_20251124` computer use tool is not supported on the Claude API and Claude Models on Google Cloud (Google managed)

882882 

models/opus-5-5/whats-new-opus-5-5 Changed · +2 / -2 lines

from line 44

4444 

4545### Thinking blocks are tied to the model and the conversation

4646 

47Every thinking block records which model produced it, and each model reads its own blocks and only some other models’. Claude Opus 5.5 reads thinking blocks from Claude Opus 5, from earlier Opus, Sonnet, and Haiku models, and, on the Claude API and Claude Models on Google Cloud (Google managed), from Claude Sonnet 5.5 and Claude Haiku 5.5, but not from Claude Fable or Claude Mythos models. On the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read thinking blocks from Claude Opus 5.5; no other model does. A conversation that moves from Claude Opus 5 onto Claude Opus 5.5, or from Claude Opus 5.5 up to Claude Fable 5.1 or Claude Mythos 5.1 on the Claude API, keeps its reasoning. One that moves from Claude Opus 5.5 to any model other than those two, or onto it from a Claude Fable or Claude Mythos model, runs the turns after the switch without the previous model’s reasoning. When a request carries a block the target model can’t read, the API drops it before the model sees it: the request succeeds, and dropped blocks aren’t billed. With the `thinking-binding-controls-2026-08-01` beta header the drop is reported in a top-level `input_transformations` array. See [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models).

47Every thinking block records which model produced it, and each model reads its own blocks and only some other models’. Claude Opus 5.5 reads thinking blocks from Claude Opus 5, from earlier Opus, Sonnet, and Haiku models, and, on the Claude API and Claude Models on Google Cloud (Google managed), from Claude Sonnet 5.5 and Claude Haiku 5.5, but not from Claude Fable or Claude Mythos models. On the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read thinking blocks from Claude Opus 5.5; no other model does. A conversation that moves from Claude Opus 5 onto Claude Opus 5.5, or from Claude Opus 5.5 up to Claude Fable 5.1 or Claude Mythos 5.1 on the Claude API, keeps its reasoning. One that moves from Claude Opus 5.5 to any model other than those two, or onto it from a Claude Fable or Claude Mythos model, runs the turns after the switch without the previous model’s reasoning. When a request carries a block the target model can’t read, the API drops it before the model sees it: the request succeeds, and dropped blocks aren’t billed. With the `thinking-binding-controls-2026-08-01` beta header, the drop is reported in a top-level `input_transformations` array. See [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models).

4848 

49The API also checks whether anything before a Claude Opus 5.5 thinking block (the `system` prompt, the `tools`, or an earlier message) has changed since the block was produced. As on Claude Fable 5.1, it enforces that check by default for accounts created on or after August 31, 2026, 00:00 UTC, on the Claude API and on cloud platforms. On those accounts, a request that replays a block after such a change returns a 400 error. To drop the affected blocks instead, send the `thinking-binding-controls-2026-08-01` beta header and set `thinking.block_binding.prefix_mismatch_behavior` to `"drop_block"`. On older accounts, setting that field to either value opts the request in. Keep the conversation append-only so the question never arises: change instructions or tools with [mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages) rather than edits. See [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking) and the migration guide’s [note on thinking blocks](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide#thinking-blocks).

49The API also checks whether anything before a Claude Opus 5.5 thinking block (the `system` prompt, the `tools`, or an earlier message) has changed since the block was produced. Keep the conversation append-only: change instructions or tools with [mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages) rather than edits. A request that replays a block after such a change and leaves `thinking.block_binding.prefix_mismatch_behavior` unset gets a 400 error on accounts created on or after August 31, 2026, 00:00 UTC. This applies on the Claude API and on cloud platforms. On accounts created before that date, the request gets the error only when it sets the field to `"error"`. To drop the affected blocks instead, send the `thinking-binding-controls-2026-08-01` beta header and set the field to `"drop_block"`. See [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking) and the migration guide’s [note on thinking blocks](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide#thinking-blocks).

5050 

5151### The `computer_20251124` computer use tool is not supported on the Claude API and Claude Models on Google Cloud (Google managed)

5252 

models/sonnet-5-5/migration-guide Changed · +1 / -1 lines

from line 841

841841 

842842Claude Sonnet 5.5 reads thinking blocks from Claude Sonnet 5, Claude Opus 4.8, Claude Haiku 4.5, and earlier models, and, on the Claude API and Claude Models on Google Cloud (Google managed), from Claude Haiku 5.5. It doesn’t read blocks from Claude Opus 5, Claude Opus 5.5, or any Claude Fable or Claude Mythos model. The API drops blocks the model can’t read. The request still returns 200, and dropped blocks aren’t billed. See [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models).

843843 

844Each Claude Sonnet 5.5 thinking block is also signed over the conversation before it. For accounts created on or after August 31, 2026, 00:00 UTC, the API enforces this by default, on the Claude API, Amazon Bedrock, and Claude Models on Google Cloud (Google managed). On those accounts, a request that replays a block after an edit to earlier history returns a 400 error. Keep conversations append-only, and change instructions or tools with [mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages). Thinking blocks that Claude Sonnet 5.5 produces work only in the account that produced them, or in an account linked to it. See [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#account-bound-thinking).

844Each Claude Sonnet 5.5 thinking block is also signed over the conversation before it. Keep conversations append-only, and change instructions or tools with [mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages). A request that replays a block after an edit to earlier history and leaves `thinking.block_binding.prefix_mismatch_behavior` unset gets a 400 error on accounts created on or after August 31, 2026, 00:00 UTC. This applies on the Claude API, Amazon Bedrock, and Claude Models on Google Cloud (Google managed). Thinking blocks that Claude Sonnet 5.5 produces work only in the account that produced them or in an account linked to it. See [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#account-bound-thinking).

845845 

846846### Computer use needs the toolset on the Claude API and Claude Models on Google Cloud (Google managed)

847847 

models/sonnet-5-5/whats-new-sonnet-5-5 Changed · +2 / -2 lines

from line 52

5252 

5353So a conversation that moves from Claude Sonnet 5 onto Claude Sonnet 5.5, or from Claude Sonnet 5.5 up to Claude Opus 5.5 on the Claude API and Claude Models on Google Cloud (Google managed), keeps its reasoning, and any other move away from Claude Sonnet 5.5 runs the turns after the switch without it. When a request carries a block the target model can’t read, the API drops it before the model sees it: the request succeeds, and dropped blocks aren’t billed. With the `thinking-binding-controls-2026-08-01` beta header, the drop is reported in a top-level `input_transformations` array. See [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models).

5454 

55The API also checks whether anything before a Claude Sonnet 5.5 thinking block has changed since the block was produced: the `system` prompt, the `tools`, or an earlier message. It enforces that check by default for accounts created on or after August 31, 2026, 00:00 UTC, on the Claude API, Amazon Bedrock, and Claude Models on Google Cloud (Google managed). On those accounts, a request that replays a block after such a change returns a 400 error. To drop the affected blocks instead, send the `thinking-binding-controls-2026-08-01` beta header and set `thinking.block_binding.prefix_mismatch_behavior` to `"drop_block"`. On older accounts, setting that field to either value opts the request in. `block_binding` works only with `thinking: {"type": "adaptive"}`. With `between_tools`, keep the history append-only, or strip the thinking blocks from the edited turn on.

55The API also checks whether anything before a Claude Sonnet 5.5 thinking block has changed since the block was produced: the `system` prompt, the `tools`, or an earlier message. Keep the conversation append-only: change instructions or tools with [mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages) rather than edits. A request that replays a block after such a change and leaves `thinking.block_binding.prefix_mismatch_behavior` unset gets a 400 error on accounts created on or after August 31, 2026, 00:00 UTC. This applies on the Claude API, Amazon Bedrock, and Claude Models on Google Cloud (Google managed). On accounts created before that date, the request gets the error only when it sets the field to `"error"`. To drop the affected blocks instead, send the `thinking-binding-controls-2026-08-01` beta header and set the field to `"drop_block"`. `block_binding` works only with `thinking: {"type": "adaptive"}`. With any other thinking setting, you can’t send `block_binding`, so strip the block the error names and every later `thinking` and `redacted_thinking` block, then retry once.

5656 

57Keep the conversation append-only so the check never fails: change instructions or tools with [mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages) rather than edits. See [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking) and the migration guide’s [note on thinking blocks](https://platform.claude.com/docs/en/models/sonnet-5-5/migration-guide#thinking-blocks).

57See [Preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking) and the migration guide’s [note on thinking blocks](https://platform.claude.com/docs/en/models/sonnet-5-5/migration-guide#thinking-blocks).

5858 

5959### The `computer_20251124` computer use tool is not supported on the Claude API and Claude Models on Google Cloud (Google managed)

6060 

test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks Changed · +1 / -1 lines

from line 204

204204 Evaluate if this query violates SEC rules, FINRA guidelines, or client privacy.

205205 ```

206206 

207 Use [structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs) to constrain the response to a Boolean classification.

207 Use [structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs) to constrain the response to a boolean classification.

208208</Accordion>

209209 

210210By layering these strategies, you create a robust defense against jailbreaking and prompt injections, ensuring your Claude-powered applications maintain the highest standards of safety and compliance.