Home Docs Claude Developer Platform

One capture · api

One read of Claude Developer Platformapi-20261028T203725Z

116 pages moved out of 821 read.

Pages moved 116 significant first

Pages read 821 in this capture

Captured 28 Oct 2026 20:37 UTC

Corpus hash 82e28663d5c2 corpus-hash

What this read moved

51-75 of 116, page 3 of 5

This capture is too large to show at once. Changes 51-75 of 116 are below, significant first; the rest are on the following screens.

build-with-claude/thinking Changed · +71 / -3 lines

## Availability ### API support #### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Data handling

from line 2

22title: Thinking

33url: https://platform.claude.com/docs/en/build-with-claude/thinking

44description: "Understand how Claude's thinking works: turn it on, read thinking output, steer thinking depth with effort, and use thinking with tools, caching, and streaming."

5featureMetadata:

6 status: ga

7 zdr:

8 eligibility: eligible

9 note: Excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements).

10 supportedModels:

11 - claude-fable-5-1

12 - claude-mythos-5-1

13 - claude-fable-5

14 - claude-mythos-5

15 - claude-mythos-preview

16 - claude-opus-5-5

17 - claude-opus-5

18 - claude-sonnet-5-5

19 - claude-sonnet-5

20 - claude-haiku-5-5

21 - claude-haiku-4-5-20251001

22 - claude-opus-4-8

23 - claude-opus-4-7

24 - claude-opus-4-6

25 - claude-sonnet-4-6

26 - claude-opus-4-5-20251101

27 - claude-sonnet-4-5-20250929

28 supportedPlatforms:

29 Claude API: ga

30 Claude Platform on AWS: ga

31 Claude Models on Google Cloud (Anthropic managed): ga

32 Amazon Bedrock: ga

33 Amazon Bedrock (InvokeModel): ga

34 Claude Models on Google Cloud (Google managed): ga

35 Microsoft Foundry: ga

536---

637

7<Note>

8 To learn how zero data retention (ZDR) applies to this feature, see [API and data retention](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention).

9</Note>

38**API support**: Usage differs by product.

1039

40* **Messages API**: On by default. Turn it off with `thinking: {"type": "disabled"}`. Where it is off by default, set `thinking` to turn it on.

41* **Managed Agents**: Always on. Managed Agents sets the thinking configuration; the agent definition has no `thinking` field.

1143A model that answers in a single pass has to get everything right on the first try: no scratch work, no checking, no changing approach halfway through. For a proof, a tricky bug, or a long agentic task, the first approach is often not the best one.

1244

1345Thinking removes that constraint. When thinking is active, Claude works through the problem in its own words before answering: it restates what is being asked, tries approaches, checks intermediate results, and abandons paths that do not hold up. That up-front thinking arrives in `thinking` content blocks ahead of the response, and Claude draws on it to produce the final answer. This is why thinking improves performance on complex tasks such as math, coding, analysis, and long-running agentic work, where the quality of the answer depends on intermediate work that would otherwise be compressed into the response itself or skipped.

from line 1280

12481280 Control how many tokens Claude uses when responding with the effort parameter, trading off between response thoroughness and token efficiency.

12491281 </Card>

12501282</CardGroup>

1283

1284## Availability

1285

1286Thinking is available on every model and eligible for zero data retention and HIPAA readiness.

1287

1288### API support

1289

1290#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

1291

1292* Availability: On by default

1293* Usage: Turn it off with `thinking: {"type": "disabled"}`. Where it is off by default, set `thinking` to turn it on.

1294

1295| Status | Cloud platforms |

1296| --------- | ----------------------- |

1297| Available | All supported platforms |

1298

1299#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

1300

1301* Availability: Always on

1302* Usage: Managed Agents sets the thinking configuration; the agent definition has no `thinking` field.

1303

1304| Status | Cloud platforms |

1305| --------- | ----------------------- |

1306| Available | All supported platforms |

1307

1308### Data handling

1309

1310[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

1311

1312| Item | Zero data retention | HIPAA |

1313| ------ | --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- |

1314| Status | Eligible (excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements)) | Eligible |

1315

1316Applicable endpoints:

1317

1318* `/v1/messages` with `thinking`

12511319

build-with-claude/token-counting Changed · +51 / -1 lines

from line 6

66 status: ga

77 zdr:

88 eligibility: eligible

9 note: Excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements).

9 note: Excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements). Count tokens before sending requests.

10 supportedModels:

11 - claude-fable-5-1

12 - claude-mythos-5-1

13 - claude-fable-5

14 - claude-mythos-5

15 - claude-mythos-preview

16 - claude-opus-5-5

17 - claude-opus-5

18 - claude-sonnet-5-5

19 - claude-sonnet-5

20 - claude-haiku-5-5

21 - claude-haiku-4-5-20251001

22 - claude-opus-4-8

23 - claude-opus-4-7

24 - claude-opus-4-6

25 - claude-sonnet-4-6

26 - claude-opus-4-5-20251101

1027 supportedPlatforms:

1128 Claude API: ga

1229 Claude Platform on AWS: ga

30 Claude Models on Google Cloud (Anthropic managed): ga

1331 Amazon Bedrock: ga

32 Amazon Bedrock (InvokeModel): ga

1433 Claude Models on Google Cloud (Google managed): ga

1534 Microsoft Foundry: ga

1635---

from line 1477

14581477 Reduce cost and latency on repeated prompts by caching prompt prefixes.

14591478 </Card>

14601479</CardGroup>

1480

1481## Availability

1482

1483Token counting is available on select models and eligible for zero data retention and HIPAA readiness.

1484

1485### API support

1486

1487#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

1488

1489* Availability: Available

1490

1491| Status | Models |

1492| --------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |

1493| Available | [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview), [Claude Mythos 5.1](https://platform.claude.com/docs/en/models/mythos-5-1/overview), [Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/overview), [Claude Mythos 5](https://platform.claude.com/docs/en/models/mythos-5/overview), Claude Mythos Preview, [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview), [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/overview), [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview), [Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/overview), [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview), [Claude Haiku 4.5](https://platform.claude.com/docs/en/models/haiku-4-5/overview), [Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-4-8/overview), [Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-4-7/overview), [Claude Opus 4.6](https://platform.claude.com/docs/en/models/opus-4-6/overview), [Claude Sonnet 4.6](https://platform.claude.com/docs/en/models/sonnet-4-6/overview), [Claude Opus 4.5](https://platform.claude.com/docs/en/models/opus-4-5/overview) |

1494

1495#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

1496

1497* Availability: Not available

1498

1499### Data handling

1500

1501[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

1502

1503| Item | Zero data retention | HIPAA |

1504| ------- | --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- |

1505| Status | Eligible (excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements)) | Eligible |

1506| Details | Count tokens before sending requests. | — |

1507

1508Applicable endpoints:

1509

1510* `/v1/messages/count_tokens`

14611511

build-with-claude/vision Changed · +60 / -1 lines

from line 2

22title: Vision

33url: https://platform.claude.com/docs/en/build-with-claude/vision

44description: Claude's vision capabilities allow it to process and analyze images, opening up exciting possibilities for multimodal interaction.

5featureMetadata:

6 status: ga

7 supportedModels:

8 - claude-fable-5-1

9 - claude-mythos-5-1

10 - claude-fable-5

11 - claude-mythos-5

12 - claude-mythos-preview

13 - claude-opus-5-5

14 - claude-opus-5

15 - claude-sonnet-5-5

16 - claude-sonnet-5

17 - claude-haiku-5-5

18 - claude-haiku-4-5-20251001

19 - claude-opus-4-8

20 - claude-opus-4-7

21 - claude-opus-4-6

22 - claude-sonnet-4-6

23 - claude-opus-4-5-20251101

24 - claude-sonnet-4-5-20250929

25 supportedPlatforms:

26 Claude API: ga

27 Claude Platform on AWS: ga

28 Claude Models on Google Cloud (Anthropic managed): ga

29 Amazon Bedrock:

30 availability: ga

31 note: Only base64 image sources; per-image limit 5 MB (base64-encoded) and a lower total request size.

32 Amazon Bedrock (InvokeModel):

33 availability: ga

34 note: Only base64 image sources; per-image limit 5 MB (base64-encoded) and a lower total request size.

35 Claude Models on Google Cloud (Google managed):

36 availability: ga

37 note: Only base64 image sources; per-image limit 5 MB (base64-encoded) and a lower total request size.

38 Microsoft Foundry: ga

539---

640

741This guide describes how to send images to Claude, the limits and costs that apply, and where to find guidance for [coordinate-based workflows](https://platform.claude.com/docs/en/build-with-claude/vision-coordinates).

from line 1283

12491283* 10 MB on [claude.ai](https://claude.ai/).

12501284

12511285<Note>

1252 Although the API supports up to 600 images for each request, [request size limits](https://platform.claude.com/docs/en/api/overview#request-size-limits) can be reached first: 32 MB for standard endpoints, and lower on some partner-operated platforms, such as Amazon Bedrock and Claude Models on Google Cloud (Google managed). For many images, consider [uploading with the Files API](https://platform.claude.com/docs/en/build-with-claude/vision#files-api-image-example) and referencing by `file_id` to keep request payloads small.

1286 Although the API supports up to 600 images for each request, [request size limits](https://platform.claude.com/docs/en/api/overview#request-size-limits) can be reached first. The limit is 128 MB on the first-party Claude API, and other platforms have their own limits. For many images, consider [uploading with the Files API](https://platform.claude.com/docs/en/build-with-claude/vision#files-api-image-example) and referencing by `file_id` to keep request payloads small.

12531287

12541288 Even when using the Files API, requests with many large images can fail before reaching the 600-image count. Reduce image dimensions or file sizes (for example, by downsampling) before uploading (see [Resolution and token cost](https://platform.claude.com/docs/en/build-with-claude/vision#evaluate-image-size)).

12551289</Note>

from line 1414

13801414 See the Messages API documentation, including example API calls involving images.

13811415 </Card>

13821416</CardGroup>

1417

1418## Availability

1419

1420Vision (images) is available on every model.

1421

1422### API support

1423

1424#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

1425

1426* Availability: Always on

1427

1428| Status | Cloud platforms |

1429| --------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

1430| Available | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws), [Claude Models on Google Cloud (Anthropic managed)](https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/anthropic-managed), [Microsoft Foundry](https://platform.claude.com/docs/en/build-with-claude/claude-in-microsoft-foundry) |

1431| Available1 | [Amazon Bedrock](https://platform.claude.com/docs/en/build-with-claude/claude-in-amazon-bedrock), [Amazon Bedrock (InvokeModel)](https://platform.claude.com/docs/en/build-with-claude/claude-on-amazon-bedrock-legacy), [Claude Models on Google Cloud (Google managed)](https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/google-managed) |

1432

14331. Only base64 image sources; per-image limit 5 MB (base64-encoded) and a lower total request size.

1434

1435#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

1436

1437* Availability: Always on

1438

1439| Status | Cloud platforms |

1440| --------- | ----------------------- |

1441| Available | All supported platforms |

13831442

build-with-claude/working-with-messages Changed · +52 / -4 lines

## Availability ### API support ### Data handling

from line 2

22title: Using the Messages API

33url: https://platform.claude.com/docs/en/build-with-claude/working-with-messages

44description: Practical patterns and examples for using the Messages API effectively

5featureMetadata:

6 status: ga

7 zdr:

8 eligibility: eligible

9 note: Excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements). Standard API calls for generating Claude responses.

10 supportedModels:

11 - claude-fable-5-1

12 - claude-mythos-5-1

13 - claude-fable-5

14 - claude-mythos-5

15 - claude-mythos-preview

16 - claude-opus-5-5

17 - claude-opus-5

18 - claude-sonnet-5-5

19 - claude-sonnet-5

20 - claude-haiku-5-5

21 - claude-haiku-4-5-20251001

22 - claude-opus-4-8

23 - claude-opus-4-7

24 - claude-opus-4-6

25 - claude-sonnet-4-6

26 - claude-opus-4-5-20251101

27 - claude-sonnet-4-5-20250929

28 supportedPlatforms:

29 Claude API: ga

30 Claude Platform on AWS: ga

31 Claude Models on Google Cloud (Anthropic managed): ga

32 Amazon Bedrock: ga

33 Amazon Bedrock (InvokeModel): ga

34 Claude Models on Google Cloud (Google managed): ga

35 Microsoft Foundry: ga

536---

637

738Anthropic offers two ways to build with Claude, each suited to different use cases:

from line 44

1344

1445This guide covers common patterns for working with the Messages API, including basic requests, multi-turn conversations, prefill techniques, and vision capabilities. For complete API specifications, see the [Messages API reference](https://platform.claude.com/docs/en/api/messages/create). For the managed agent harness instead, see the [Claude Managed Agents overview](https://platform.claude.com/docs/en/managed-agents/overview).

1546

16<Note>

17 To learn how zero data retention (ZDR) applies to this feature, see [API and data retention](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention).

18</Note>

2047## Basic request and response

2148

2249<Note>

from line 1082

10551082 Set an advisory token budget across a full agentic loop with `output_config.task_budget`.

10561083 </Card>

10571084</CardGroup>

1085

1086## Availability

1087

1088The Messages API is available on every model and eligible for zero data retention and HIPAA readiness.

1089

1090### API support

1091

1092Available

1093

1094### Data handling

1095

1096[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

1097

1098| Item | Zero data retention | HIPAA |

1099| ------- | --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- |

1100| Status | Eligible (excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements)) | Eligible |

1101| Details | Standard API calls for generating Claude responses. | — |

1102

1103Applicable endpoints:

1104

1105* `/v1/messages`

10581106

cli-sdks-libraries/libraries/apple-foundation-models Changed · +36 / -0 lines

## Availability ### API support #### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

from line 2

22title: Apple Foundation Models

33url: https://platform.claude.com/docs/en/cli-sdks-libraries/libraries/apple-foundation-models

44description: Use Claude on Apple platforms through the Foundation Models framework with the Claude for Foundation Models Swift package.

5featureMetadata:

6 status: beta

7 supportedModels:

8 - claude-fable-5-1

9 - claude-mythos-5-1

10 - claude-fable-5

11 - claude-mythos-5

12 - claude-mythos-preview

13 - claude-opus-5-5

14 - claude-opus-5

15 - claude-sonnet-5-5

16 - claude-sonnet-5

17 - claude-haiku-5-5

18 - claude-haiku-4-5-20251001

19 - claude-opus-4-8

20 - claude-opus-4-7

21 - claude-opus-4-6

22 - claude-sonnet-4-6

23 - claude-opus-4-5-20251101

24 - claude-sonnet-4-5-20250929

25 supportedPlatforms:

26 Claude API: beta

527---

628

729[Claude for Foundation Models](https://github.com/anthropics/ClaudeForFoundationModels) is a Swift package that makes Claude available as a server-side language model in Apple’s [Foundation Models](https://developer.apple.com/documentation/foundationmodels) framework. The package conforms Claude to the framework’s `LanguageModel` protocol, so you drive it with the same `LanguageModelSession` API you use for Apple’s on-device model: `respond(to:)`, streaming, guided generation, and tool calling all work the same way.

from line 285

263285| [Claude API reference](https://platform.claude.com/docs/en/api/overview) | The underlying Messages API |

264286

265287The package is licensed under Apache 2.0. Bug reports are welcome through GitHub issues. External pull requests are not being accepted during the beta period.

288

289## Availability

290

291Apple Foundation Models are in beta on every model.

292

293### API support

294

295#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

296

297* Availability: Available

298

299| Status | Cloud platforms |

300| ------ | --------------- |

301| Beta | Claude API |

266302

cli-sdks-libraries/libraries/openai-sdk Changed · +36 / -0 lines

## Availability ### API support #### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

from line 2

22title: OpenAI SDK compatibility

33url: https://platform.claude.com/docs/en/cli-sdks-libraries/libraries/openai-sdk

44description: Anthropic provides a compatibility layer that lets you use the OpenAI SDK to test the Claude API. With a few code changes, you can evaluate Anthropic model capabilities.

5featureMetadata:

6 status: ga

7 supportedModels:

8 - claude-fable-5-1

9 - claude-mythos-5-1

10 - claude-fable-5

11 - claude-mythos-5

12 - claude-mythos-preview

13 - claude-opus-5-5

14 - claude-opus-5

15 - claude-sonnet-5-5

16 - claude-sonnet-5

17 - claude-haiku-5-5

18 - claude-haiku-4-5-20251001

19 - claude-opus-4-8

20 - claude-opus-4-7

21 - claude-opus-4-6

22 - claude-sonnet-4-6

23 - claude-opus-4-5-20251101

24 - claude-sonnet-4-5-20250929

25 supportedPlatforms:

26 Claude API: ga

527---

628

729<Note>

from line 496

474496| `openai-version` | Always `2020-10-01` |

475497| `authorization` | Fully supported |

476498| `openai-processing-ms` | Always empty |

499

500## Availability

501

502OpenAI SDK compatibility is available on every model.

503

504### API support

505

506#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

507

508* Availability: Available

509

510| Status | Cloud platforms |

511| --------- | --------------- |

512| Available | Claude API |

477513

cli-sdks-libraries/middleware Changed · +49 / -0 lines

from line 2

22title: SDK middleware

33url: https://platform.claude.com/docs/en/cli-sdks-libraries/middleware

44description: Intercept and modify the SDK's requests and responses.

5featureMetadata:

6 supportedModels:

7 - claude-fable-5-1

8 - claude-mythos-5-1

9 - claude-fable-5

10 - claude-mythos-5

11 - claude-mythos-preview

12 - claude-opus-5-5

13 - claude-opus-5

14 - claude-sonnet-5-5

15 - claude-sonnet-5

16 - claude-haiku-5-5

17 - claude-haiku-4-5-20251001

18 - claude-opus-4-8

19 - claude-opus-4-7

20 - claude-opus-4-6

21 - claude-sonnet-4-6

22 - claude-opus-4-5-20251101

23 - claude-sonnet-4-5-20250929

24 supportedPlatforms:

25 Claude API: ga

26 Claude Platform on AWS: ga

27 Amazon Bedrock: ga

28 Amazon Bedrock (InvokeModel): ga

29 Claude Models on Google Cloud (Google managed): ga

30 Microsoft Foundry: ga

531---

632

733The Claude SDK provides a middleware (or interceptor) hook that lets you run code before a request is sent and after the response is received. Use middleware for cross-cutting concerns such as logging, custom retries, request annotation, and refusal fallback handling.

from line 208

182208## Built-in middleware

183209

184210The SDK ships a refusal-fallback middleware that automatically retries a request Claude declines (`stop_reason: "refusal"`) on a fallback model. [Refusals and fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback) lists the models whose safety classifiers can decline requests, and [Detect and retry on a fallback model](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#client-side-fallback) covers setup and per-language examples.

211

212## Availability

213

214SDK middleware is available or in beta, depending on the API.

215

216### API support

217

218#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

219

220* Availability: Available

221

222| Status | Cloud platforms |

223| --------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

224| Available | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws), [Amazon Bedrock](https://platform.claude.com/docs/en/build-with-claude/claude-in-amazon-bedrock), [Amazon Bedrock (InvokeModel)](https://platform.claude.com/docs/en/build-with-claude/claude-on-amazon-bedrock-legacy), [Claude Models on Google Cloud (Google managed)](https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/google-managed), [Microsoft Foundry](https://platform.claude.com/docs/en/build-with-claude/claude-in-microsoft-foundry) |

225

226#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

227

228* Availability: Available

229* Usage: Runs on your SDK client’s calls to the Managed Agents API. It does not see the model calls the agent makes inside its loop.

230

231| Status | Cloud platforms |

232| ------ | ------------------------------------------------------------------------------------------------------------------ |

233| Beta | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) |

185234

manage-claude/api-and-data-retention Changed · +59 / -40 lines

### Platform features ### Messages API features ### Managed Agents features ### Other features

from line 159

159159

160160## Feature eligibility

161161

162The following table lists which Claude API features are eligible for ZDR and HIPAA readiness arrangements.

162The following tables list which Claude API features are eligible for ZDR and HIPAA readiness arrangements. They split the features as the [Features overview](https://platform.claude.com/docs/en/build-with-claude/overview) does.

163163

164164Each eligibility column uses three values:

165165

from line 167

167167* **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.

168168* **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.

169169

171| -------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------- | ------------------------------------------------------- | ----------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

174| [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. |

175| [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 according to standard policy. See [Agent Skills data retention](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview#data-retention). |

177| [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 data retention](https://platform.claude.com/docs/en/build-with-claude/batch-processing#data-retention). |

178| [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 data retention](https://platform.claude.com/docs/en/agents-and-tools/tool-use/browser-use-tool#data-retention). |

179| [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 allow comparison against the next request. See [Cache diagnostics data retention](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics#data-retention). |

181| [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). |

182| [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 data retention](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool#data-retention). |

183| [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 data retention](https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool#data-retention). |

184| [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. |

185| [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. |

188| [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| [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 file lifecycle](https://platform.claude.com/docs/en/build-with-claude/files#data-retention). |

191| [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 according to standard policy. See [MCP connector data retention](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector#data-retention). |

192| [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. |

195| [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. |

196| [OpenTelemetry export](https://platform.claude.com/docs/en/manage-claude/opentelemetry) | `/v1/organizations/telemetry_destinations` | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Beta. Spans have the same ZDR eligibility as the feature that produces them. HIPAA organizations can export metadata. SDK spans go directly from your environment to your exporter. After delivery, the data is held by the service that you chose, outside Anthropic, under that service’s retention and access controls. |

198| [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 data retention](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling#data-retention). |

199| [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). |

201| [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 data retention](https://platform.claude.com/docs/en/build-with-claude/structured-outputs#data-retention). |

205| [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 run by Anthropic; the tool definitions in the request are searched in memory per call and nothing is stored after the response. |

206| [User profiles](https://platform.claude.com/docs/en/manage-claude/user-profiles) | `/v1/user_profiles` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Profiles are metadata about your end customers and don’t contain prompts or outputs. The profile data you send is retained according to standard policy until you delete the profile. HIPAA-enabled organizations can use user profiles, but the user profiles API is not a HIPAA-eligible service and PHI must not be included in the profile data you send. See [What a profile contains and how Anthropic uses it](https://platform.claude.com/docs/en/manage-claude/user-profiles#what-a-profile-contains-and-how-anthropic-uses-it). |

207| [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 for web fetch](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 might retain request data (such as fetched URLs and request metadata) according to their own policies. |

208| [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 for web search](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool#dynamic-filtering) is not eligible for ZDR or HIPAA. |

170### Platform features

171

172Features that relate to both the [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) and [Claude Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview). Some are available in only one of them. This table gives each feature’s eligibility in the Messages API. In Claude Managed Agents, no feature is eligible for ZDR or HIPAA readiness; see the Claude Managed Agents row under [Managed Agents features](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#managed-agents-features).

173

175| :------------------------------------------------------------------------------------------------------------------------- | :--------------------------------------------- | :-------------- | :------------- | :---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

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

177| [Agent Skills](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview) | `/v1/skills`, `/v1/messages` with `skills` | No | No | Skill data retained according to standard policy. [Learn more](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview#data-retention) |

178| [Batch processing](https://platform.claude.com/docs/en/build-with-claude/batch-processing) | `/v1/messages/batches` | No | No | 29-day retention; async storage required. [Learn more](https://platform.claude.com/docs/en/build-with-claude/batch-processing#data-retention) |

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

180| [Citations](https://platform.claude.com/docs/en/build-with-claude/citations) | `/v1/messages` | Yes | Yes | |

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

182| [Data residency](https://platform.claude.com/docs/en/manage-claude/data-residency) | `/v1/messages` with `inference_geo` | Yes | Yes | |

183| [Effort](https://platform.claude.com/docs/en/build-with-claude/effort) | `/v1/messages` with `effort` | Yes | Yes | |

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

185| [Files API](https://platform.claude.com/docs/en/build-with-claude/files) | `/v1/files` | No | No | Files retained until explicitly deleted or they reach their configured expiration. [Learn more](https://platform.claude.com/docs/en/build-with-claude/files#data-retention) |

186| [MCP connector](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector) | `/v1/messages` with `mcp_servers` | No | No | Data retained according to standard policy. [Learn more](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector#data-retention) |

187| [MCP tunnels](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/overview) | `/v1/tunnels` | No | No | Research preview. The MCP tunnels security page describes the data-flow boundary and subprocessor details. [Learn more](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/security) |

188| [Mid-conversation system messages](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages) | `/v1/messages` with `role: "system"` messages | Yes | Yes | 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. |

189| [PDF support](https://platform.claude.com/docs/en/build-with-claude/pdf-support) | `/v1/messages` | Yes | Yes | HIPAA eligibility applies to PDFs sent inline through the Messages API, not through the Files API. |

190| [Programmatic tool calling](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling) | `/v1/messages` with `code_execution` tool | No | No | Built on code execution containers; data retained up to 30 days. [Learn more](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling#data-retention) |

191| [Prompt caching](https://platform.claude.com/docs/en/build-with-claude/prompt-caching) | `/v1/messages` | Yes | Yes | 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. [Learn more](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#data-retention) |

192| ↳ [Cache diagnostics](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics) | `/v1/messages` with `diagnostics` | Yes (qualified) | No | Your prompts and Claude’s outputs are not stored. A fingerprint of cryptographic hashes and token-count estimates is retained briefly to allow comparison against the next request. [Learn more](https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics#data-retention) |

193| [Search results](https://platform.claude.com/docs/en/build-with-claude/search-results) | `/v1/messages` with `search_results` source | Yes | Yes | |

194| [Structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs) | `/v1/messages` | Yes (qualified) | Yes | 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 (`strict: true` on tools), which uses the same grammar pipeline. [Learn more](https://platform.claude.com/docs/en/build-with-claude/structured-outputs#data-retention) PHI must not be included in JSON schema definitions; see PHI handling guidelines. [Learn more](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#phi-handling-guidelines) |

195| [Thinking](https://platform.claude.com/docs/en/build-with-claude/thinking) | `/v1/messages` with `thinking` | Yes | Yes | |

196| ↳ [Adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking-steering-and-cost) | `/v1/messages` | Yes | Yes | |

197| [Threshold compaction](https://platform.claude.com/docs/en/build-with-claude/compaction-threshold) | `/v1/messages` with `context_management` | Yes | No | Server-side compaction results are returned and round-tripped statelessly through the API response. |

198| [Token counting](https://platform.claude.com/docs/en/build-with-claude/token-counting) | `/v1/messages/count_tokens` | Yes | Yes | Count tokens before sending requests. |

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

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

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

202

203### Messages API features

204

206| :----------------------------------------------------------------------------------------------------------------------- | :---------------------------------------- | :----------- | :------------- | :---------------------------------------------------------------------------------------------------------------------------------------------------- |

207| [Bash tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/bash-tool) | `/v1/messages` with `bash` tool | Yes | Yes | Client-side tool run in your environment. |

208| [Code execution tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool) | `/v1/messages` with `code_execution` tool | No | No | Container data retained up to 30 days. [Learn more](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool#data-retention) |

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

210| [Fine-grained tool streaming](https://platform.claude.com/docs/en/agents-and-tools/tool-use/fine-grained-tool-streaming) | `/v1/messages` | Yes | Yes | |

211| [Memory tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/memory-tool) | `/v1/messages` with `memory` tool | Yes | Yes | Client-side memory storage where you control data retention. |

212| [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) | `/v1/messages` | Yes | Yes | Standard API calls for generating Claude responses. |

213| [Text editor tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/text-editor-tool) | `/v1/messages` with `text_editor` tool | Yes | Yes | Client-side tool run in your environment. |

214

215### Managed Agents features

216

218| :----------------------------------------------------------------------------------- | :----------------------------------------------- | :----------- | :------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------- |

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

220

221### Other features

222

224| ------------------------------------------------------------------------------------------------ | ------------------------------------------ | ----------------------------------- | ----------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

226| [OpenTelemetry export](https://platform.claude.com/docs/en/manage-claude/opentelemetry) | `/v1/organizations/telemetry_destinations` | <Eligible>Yes</Eligible> | <Eligible status="no">No</Eligible> | Beta. Spans have the same ZDR eligibility as the feature that produces them. HIPAA organizations can export metadata. SDK spans go directly from your environment to your exporter. After delivery, the data is held by the service that you chose, outside Anthropic, under that service’s retention and access controls. |

227| [User profiles](https://platform.claude.com/docs/en/manage-claude/user-profiles) | `/v1/user_profiles` | <Eligible status="no">No</Eligible> | <Eligible status="no">No</Eligible> | Profiles are metadata about your end customers and don’t contain prompts or outputs. The profile data you send is retained according to standard policy until you delete the profile. HIPAA-enabled organizations can use user profiles, but the user profiles API is not a HIPAA-eligible service and PHI must not be included in the profile data you send. See [What a profile contains and how Anthropic uses it](https://platform.claude.com/docs/en/manage-claude/user-profiles#what-a-profile-contains-and-how-anthropic-uses-it). |

209228

210229## Retention regardless of arrangement

211230

manage-claude/data-residency Changed · +76 / -4 lines

from line 2

22title: Data residency

33url: https://platform.claude.com/docs/en/manage-claude/data-residency

44description: Manage where model inference runs and where data is stored with geographic controls.

5featureMetadata:

6 status: ga

7 zdr:

8 eligibility: eligible

9 note: Excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements).

10 supportedModels:

11 - claude-fable-5-1

12 - claude-mythos-5-1

13 - claude-fable-5

14 - claude-mythos-5

15 - claude-mythos-preview

16 - claude-opus-5-5

17 - claude-opus-5

18 - claude-sonnet-5-5

19 - claude-sonnet-5

20 - claude-haiku-5-5

21 - claude-opus-4-8

22 - claude-opus-4-7

23 - claude-opus-4-6

24 - claude-sonnet-4-6

25 supportedPlatforms:

26 Claude API: ga

27 Claude Platform on AWS:

28 availability: ga

29 note: Workspace geo is not configurable; Managed Agents sessions there run with workspace geo "us".

30 Claude Models on Google Cloud (Anthropic managed):

31 availability: ga

32 note: Workspace geo is not configurable.

533---

634

35**API support**: Usage differs by product.

37* **Messages API**: Set `inference_geo` on each request, or set a default for the workspace.

38* **Managed Agents**: Pin `inference_geo` in the agent’s model configuration; override it at session creation.

740Data residency controls let you manage where your data is processed and stored. Two independent settings govern this:

841

942* **Inference geo:** Controls where model inference runs, on a per-request basis. Set through the `inference_geo` API parameter or as a workspace default.

from line 48

1548

1649## Inference geo

1750

18<Note>

19 To learn how zero data retention (ZDR) applies to this feature, see [API and data retention](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention).

20</Note>

2251The `inference_geo` parameter controls where model inference runs for a specific API request. Add it to any `POST /v1/messages` call.

2352

2453| Value | Description |

from line 356

327356 Track usage and costs by data residency.

328357 </Card>

329358</CardGroup>

359

360## Availability

361

362Data residency is available on select models and eligible for zero data retention and HIPAA readiness.

363

364### API support

365

366#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

367

368* Availability: Available

369* Usage: Set `inference_geo` on each request, or set a default for the workspace.

370

371| Status | Cloud platforms | Models |

372| --------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

373| Available | Claude API | [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview), [Claude Mythos 5.1](https://platform.claude.com/docs/en/models/mythos-5-1/overview), [Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/overview), [Claude Mythos 5](https://platform.claude.com/docs/en/models/mythos-5/overview), Claude Mythos Preview, [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview), [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/overview), [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview), [Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/overview), [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview), [Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-4-8/overview), [Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-4-7/overview), [Claude Opus 4.6](https://platform.claude.com/docs/en/models/opus-4-6/overview), [Claude Sonnet 4.6](https://platform.claude.com/docs/en/models/sonnet-4-6/overview) |

374| Available1 | [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) | [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview), [Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/overview), [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview), [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/overview), [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview), [Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/overview), [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview), [Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-4-8/overview), [Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-4-7/overview), [Claude Opus 4.6](https://platform.claude.com/docs/en/models/opus-4-6/overview), [Claude Sonnet 4.6](https://platform.claude.com/docs/en/models/sonnet-4-6/overview) |

375| Available1 | [Claude Models on Google Cloud (Anthropic managed)](https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/anthropic-managed) | [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview), [Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/overview), [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview), [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/overview), [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview), [Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/overview), [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview), [Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-4-8/overview), [Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-4-7/overview), [Claude Opus 4.6](https://platform.claude.com/docs/en/models/opus-4-6/overview), [Claude Sonnet 4.6](https://platform.claude.com/docs/en/models/sonnet-4-6/overview) |

376

3771. Workspace geo is not configurable.

378

379#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

380

381* Availability: Available

382* Usage: Pin `inference_geo` in the agent’s model configuration, and override it per session at session creation. An agent with no pin follows the workspace default.

383

384| Status | Cloud platforms | Models |

385| --------------------- | ------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

386| Available | Claude API | [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview), [Claude Mythos 5.1](https://platform.claude.com/docs/en/models/mythos-5-1/overview), [Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/overview), [Claude Mythos 5](https://platform.claude.com/docs/en/models/mythos-5/overview), Claude Mythos Preview, [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview), [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/overview), [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview), [Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/overview), [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview), [Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-4-8/overview), [Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-4-7/overview), [Claude Opus 4.6](https://platform.claude.com/docs/en/models/opus-4-6/overview), [Claude Sonnet 4.6](https://platform.claude.com/docs/en/models/sonnet-4-6/overview) |

387| Available1 | [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) | [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview), [Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/overview), [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview), [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/overview), [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview), [Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/overview), [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview), [Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-4-8/overview), [Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-4-7/overview), [Claude Opus 4.6](https://platform.claude.com/docs/en/models/opus-4-6/overview), [Claude Sonnet 4.6](https://platform.claude.com/docs/en/models/sonnet-4-6/overview) |

388

3891. Workspace geo is not configurable; Managed Agents sessions there run with workspace geo “us”.

390

391### Data handling

392

393[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

394

395| Item | Zero data retention | HIPAA |

396| ------ | --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- |

397| Status | Eligible (excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements)) | Eligible |

398

399Applicable endpoints:

400

401* `/v1/messages` with `inference_geo`

330402

manage-claude/opentelemetry Changed · +48 / -0 lines

from line 7

77 title: OpenTelemetry

88 url: https://platform.claude.com/docs/en/manage-claude/opentelemetry

99 status: beta

10 supportedModels:

11 - claude-fable-5-1

12 - claude-mythos-5-1

13 - claude-fable-5

14 - claude-mythos-5

15 - claude-mythos-preview

16 - claude-opus-5-5

17 - claude-opus-5

18 - claude-sonnet-5-5

19 - claude-sonnet-5

20 - claude-haiku-5-5

21 - claude-haiku-4-5-20251001

22 - claude-opus-4-8

23 - claude-opus-4-7

24 - claude-opus-4-6

25 - claude-sonnet-4-6

26 - claude-opus-4-5-20251101

27 - claude-sonnet-4-5-20250929

28 supportedPlatforms:

29 Claude API: beta

30 Claude Platform on AWS: beta

31 Claude Models on Google Cloud (Anthropic managed): beta

32 Amazon Bedrock: beta

33 Amazon Bedrock (InvokeModel): beta

34 Claude Models on Google Cloud (Google managed): beta

35 Microsoft Foundry: beta

1036---

1137

1238<Note>

from line 263

237263<Note>

238264 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.

239265</Note>

266

267## Availability

268

269OpenTelemetry is in beta on every model.

270

271### API support

272

273#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

274

275* Availability: Available

276

277| Status | Cloud platforms | Models |

278| ------ | ----------------------- | ----------------------------------------------------------------- |

279| Beta | All supported platforms | [All models](https://platform.claude.com/docs/en/models/overview) |

280

281#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

282

283* Availability: Available

284

285| Status | Cloud platforms | Models |

286| ------ | ----------------------- | ----------------------------------------------------------------- |

287| Beta | All supported platforms | [All models](https://platform.claude.com/docs/en/models/overview) |

240288

manage-claude/opentelemetry/api Changed · +58 / -6 lines

## Zero data retention ## Availability ### API support #### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Data handling

from line 9

99 status: beta

1010 zdr:

1111 eligibility: eligible

12 note: Excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements). Only for features that are themselves ZDR eligible, such as the Messages API. Claude Managed Agents isn't ZDR eligible.

12 note: Excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements). Beta. Spans have the same ZDR eligibility as the feature that produces them. HIPAA organizations can export metadata. SDK spans go directly from your environment to your exporter. After delivery, the data is held by the service that you chose, outside Anthropic, under that service's retention and access controls.

13 supportedModels:

14 - claude-fable-5-1

15 - claude-mythos-5-1

16 - claude-fable-5

17 - claude-mythos-5

18 - claude-mythos-preview

19 - claude-opus-5-5

20 - claude-opus-5

21 - claude-sonnet-5-5

22 - claude-sonnet-5

23 - claude-haiku-5-5

24 - claude-haiku-4-5-20251001

25 - claude-opus-4-8

26 - claude-opus-4-7

27 - claude-opus-4-6

28 - claude-sonnet-4-6

29 - claude-opus-4-5-20251101

30 - claude-sonnet-4-5-20250929

1331 supportedPlatforms:

1432 Claude API: beta

1533 Claude Platform on AWS: beta

16 rows:

17 - label: Works with

18 value: Messages API, Claude Managed Agents

19 details:

20 - After delivery, the data is held by the service that you chose, outside Anthropic, under that service's retention and access controls.

2134---

2235

2336With OpenTelemetry, the Claude API can emit a span for each model request and each tool use that it runs. You can view these spans in your observability platform without changing your code.

from line 688

675688| Length of all header names and values together | 16,384 bytes |

676689| Length of an endpoint URL | 2,048 characters |

677690

691## Zero data retention

692

693Spans have the same ZDR eligibility as the feature that produces them. Claude Managed Agents isn’t ZDR eligible. See [Feature eligibility](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility).

694

678695## Troubleshooting

679696

680697| Problem | What to check |

from line 717

700717| Redirect (status 3xx) | The endpoint redirected the request. The API doesn’t follow redirects, so use the URL it redirects to |

701718| HTTP error (status 408, 429, or 5xx) | The endpoint was overloaded, limited the rate of requests, or had an error |

702719| HTTP error (another 4xx status) | The endpoint refused the request, for example because of its size or format |

720

721## Availability

722

723API tracing is in beta on every model and eligible for zero data retention.

724

725### API support

726

727#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

728

729* Availability: Available

730

731| Status | Cloud platforms |

732| ------ | ------------------------------------------------------------------------------------------------------------------ |

733| Beta | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) |

734

735#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

736

737* Availability: Available

738

739| Status | Cloud platforms |

740| ------ | ------------------------------------------------------------------------------------------------------------------ |

741| Beta | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) |

742

743### Data handling

744

745[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

746

747| Item | Zero data retention | HIPAA |

748| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------- |

749| Status | Eligible (excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements)) | Not eligible |

750| Details | Beta. Spans have the same ZDR eligibility as the feature that produces them. HIPAA organizations can export metadata. SDK spans go directly from your environment to your exporter. After delivery, the data is held by the service that you chose, outside Anthropic, under that service’s retention and access controls. | HIPAA organizations can export metadata. |

751

752Applicable endpoints:

753

754* `/v1/organizations/telemetry_destinations`

703755

manage-claude/opentelemetry/sdk Changed · +42 / -3 lines

from line 7

77 title: OpenTelemetry

88 url: https://platform.claude.com/docs/en/manage-claude/opentelemetry

99 status: beta

10 supportedModels:

11 - claude-fable-5-1

12 - claude-mythos-5-1

13 - claude-fable-5

14 - claude-mythos-5

15 - claude-mythos-preview

16 - claude-opus-5-5

17 - claude-opus-5

18 - claude-sonnet-5-5

19 - claude-sonnet-5

20 - claude-haiku-5-5

21 - claude-haiku-4-5-20251001

22 - claude-opus-4-8

23 - claude-opus-4-7

24 - claude-opus-4-6

25 - claude-sonnet-4-6

26 - claude-opus-4-5-20251101

27 - claude-sonnet-4-5-20250929

1028 supportedPlatforms:

1129 Claude API: beta

1230 Claude Platform on AWS: beta

1331 Amazon Bedrock: beta

32 Amazon Bedrock (InvokeModel): beta

1433 Claude Models on Google Cloud (Google managed): beta

1534 Microsoft Foundry: beta

16 rows:

17 - label: Works with

18 value: Messages API, Claude Managed Agents

1935---

2036

2137With OpenTelemetry, the Claude SDK can emit a span for each request that it sends and each tool use that the [tool runner](https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-runner) runs. You can view these spans in your observability platform, nested in your own traces.

from line 673

657673| Requests from the SDK don’t have a `traceparent` header | Register the W3C trace context propagator with OpenTelemetry. The SDK writes the headers with your application’s propagator. |

658674| A stream’s SDK span doesn’t end or doesn’t have the response | See [Tracing a stream](https://platform.claude.com/docs/en/manage-claude/opentelemetry/sdk#trace-a-stream). |

659675| Each request appears twice | A request’s SDK span and API span have the same name. Use the span kind or `service.name` to tell them apart. If another library also traces the SDK, turn off one of the two. |

676

677## Availability

678

679SDK tracing is in beta on every model.

680

681### API support

682

683#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

684

685* Availability: Available

686

687| Status | Cloud platforms |

688| ------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

689| Beta | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws), [Amazon Bedrock](https://platform.claude.com/docs/en/build-with-claude/claude-in-amazon-bedrock), [Amazon Bedrock (InvokeModel)](https://platform.claude.com/docs/en/build-with-claude/claude-on-amazon-bedrock-legacy), [Claude Models on Google Cloud (Google managed)](https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/google-managed), [Microsoft Foundry](https://platform.claude.com/docs/en/build-with-claude/claude-in-microsoft-foundry) |

690

691#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

692

693* Availability: Available

694* Usage: The SDK traces the requests that your code sends, such as the events you send to a session; spans for the agent’s model requests and tool calls come from API tracing. In a self-hosted sandbox, the SDK’s environment worker also traces each tool it runs.

695

696| Status | Cloud platforms |

697| ------ | ------------------------------------------------------------------------------------------------------------------ |

698| Beta | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) |

660699

managed-agents/agent-setup Changed · +58 / -1 lines

## Availability ### API support #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Required beta headers ### Data handling

from line 6

66 topic:

77 title: Managed Agents

88 url: https://platform.claude.com/docs/en/managed-agents/overview

9 status: beta

109 betaHeader: managed-agents-2026-04-01

10 zdr:

11 eligibility: not-eligible

12 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

13 supportedModels:

14 - claude-fable-5-1

15 - claude-mythos-5-1

16 - claude-fable-5

17 - claude-mythos-5

18 - claude-mythos-preview

19 - claude-opus-5-5

20 - claude-opus-5

21 - claude-sonnet-5-5

22 - claude-sonnet-5

23 - claude-haiku-5-5

24 - claude-haiku-4-5-20251001

25 - claude-opus-4-8

26 - claude-opus-4-7

27 - claude-opus-4-6

28 - claude-sonnet-4-6

29 - claude-opus-4-5-20251101

30 - claude-sonnet-4-5-20250929

1131---

1232

1333An agent is a reusable, versioned configuration that defines persona and capabilities. It bundles the model, system prompt, tools, MCP servers, and skills that shape how Claude behaves during a session.

from line 857

837857 Event types, self-hosted worker CLI flags, supported MCP server types, rate limits, and branding guidelines for Claude Managed Agents.

838858 </Card>

839859</CardGroup>

860

861## Availability

862

863Agent configurations are available on every model.

864

865### API support

866

867#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

868

869* Availability: Available

870

871| Status | Cloud platforms | Models |

872| --------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------- |

873| Available | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws), [Claude Models on Google Cloud (Anthropic managed)](https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/anthropic-managed) | [All models](https://platform.claude.com/docs/en/models/overview) |

874

875### Required beta headers

876

877[Beta headers overview](https://platform.claude.com/docs/en/api/beta-headers)

878

879Beta headers allow you to access experimental features and new model capabilities before they become part of the standard API.

880

881* `managed-agents-2026-04-01`

882

883### Data handling

884

885[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

886

887| Item | Zero data retention | HIPAA |

888| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |

889| Status | Not eligible | Not eligible |

890| Details | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes. | Applies to all Managed Agents sub-features, including Self-hosted sandboxes. |

891

892Applicable endpoints:

893

894* `/v1/agents`

895* `/v1/sessions`

896* `/v1/environments`

840897

managed-agents/budgets Changed · +67 / -0 lines

from line 8

88 url: https://platform.claude.com/docs/en/managed-agents/overview

99 status: beta

1010 betaHeader: managed-agents-2026-04-01

11 zdr:

12 eligibility: not-eligible

13 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

14 supportedModels:

15 - claude-fable-5-1

16 - claude-mythos-5-1

17 - claude-fable-5

18 - claude-mythos-5

19 - claude-mythos-preview

20 - claude-opus-5-5

21 - claude-opus-5

22 - claude-sonnet-5-5

23 - claude-sonnet-5

24 - claude-haiku-5-5

25 - claude-haiku-4-5-20251001

26 - claude-opus-4-8

27 - claude-opus-4-7

28 - claude-opus-4-6

29 - claude-sonnet-4-6

30 - claude-opus-4-5-20251101

31 - claude-sonnet-4-5-20250929

32 supportedPlatforms:

33 Claude API: beta

34 Claude Platform on AWS: beta

35 Claude Models on Google Cloud (Anthropic managed): beta

1136---

1237

1338A session budget is an optional hard spend ceiling you set when you [create a session](https://platform.claude.com/docs/en/managed-agents/sessions). The platform continuously prices everything the session consumes at public list rates (the session’s **list cost**) and stops issuing new model requests once that cost reaches the budget. The request in flight when the cap is crossed still completes, so the final list cost can land [a fraction past the budget](https://platform.claude.com/docs/en/managed-agents/budgets#when-a-session-reaches-its-budget). A session at its budget pauses and goes [idle](https://platform.claude.com/docs/en/managed-agents/session-operations#session-statuses) rather than terminating; changing or removing the budget resumes the work that the budget paused. Deployments accept the same budget and apply it to each session they start; see [Budgets on deployments](https://platform.claude.com/docs/en/managed-agents/budgets#budgets-on-deployments).

from line 451

426451<Note>

427452 Session budgets are hard caps in US dollars (written in cents) on a single session, enforced by the platform. They are distinct from the Messages API’s [task budgets](https://platform.claude.com/docs/en/build-with-claude/task-budgets), which are advisory, token-denominated budgets the model uses to self-regulate within one agentic loop.

428453</Note>

454

455## Availability

456

457Session budgets are in beta on every model.

458

459### API support

460

461#### [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages)

462

463* Availability: Not available

464* Usage: Use [Task budgets](https://platform.claude.com/docs/en/build-with-claude/task-budgets) instead.

465

466#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

467

468* Availability: Available

469

470| Status | Cloud platforms |

471| ------ | ----------------------- |

472| Beta | All supported platforms |

473

474### Required beta headers

475

476[Beta headers overview](https://platform.claude.com/docs/en/api/beta-headers)

477

478Beta headers allow you to access experimental features and new model capabilities before they become part of the standard API.

479

480* `managed-agents-2026-04-01`

481

482### Data handling

483

484[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

485

486| Item | Zero data retention | HIPAA |

487| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |

488| Status | Not eligible | Not eligible |

489| Details | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes. | Applies to all Managed Agents sub-features, including Self-hosted sandboxes. |

490

491Applicable endpoints:

492

493* `/v1/agents`

494* `/v1/sessions`

495* `/v1/environments`

429496

managed-agents/choosing-an-environment Changed · +25 / -0 lines

from line 8

88 url: https://platform.claude.com/docs/en/managed-agents/overview

99 status: beta

1010 betaHeader: managed-agents-2026-04-01

11 zdr:

12 eligibility: not-eligible

13 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

14 supportedModels:

15 - claude-fable-5-1

16 - claude-mythos-5-1

17 - claude-fable-5

18 - claude-mythos-5

19 - claude-mythos-preview

20 - claude-opus-5-5

21 - claude-opus-5

22 - claude-sonnet-5-5

23 - claude-sonnet-5

24 - claude-haiku-5-5

25 - claude-haiku-4-5-20251001

26 - claude-opus-4-8

27 - claude-opus-4-7

28 - claude-opus-4-6

29 - claude-sonnet-4-6

30 - claude-opus-4-5-20251101

31 - claude-sonnet-4-5-20250929

32 supportedPlatforms:

33 Claude API: beta

34 Claude Platform on AWS: beta

35 Claude Models on Google Cloud (Anthropic managed): beta

1136---

1237

1338An environment sets where your agent’s tools and code run. They run in a sandbox. A sandbox is an isolated workspace, such as a container, where the agent runs commands and works on files. You have two options:

managed-agents/cloud-sandboxes-reference Changed · +21 / -1 lines

from line 6

66 topic:

77 title: Managed Agents

88 url: https://platform.claude.com/docs/en/managed-agents/overview

9 status: beta

109 betaHeader: managed-agents-2026-04-01

10 zdr:

11 eligibility: not-eligible

12 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

13 supportedModels:

14 - claude-fable-5-1

15 - claude-mythos-5-1

16 - claude-fable-5

17 - claude-mythos-5

18 - claude-mythos-preview

19 - claude-opus-5-5

20 - claude-opus-5

21 - claude-sonnet-5-5

22 - claude-sonnet-5

23 - claude-haiku-5-5

24 - claude-haiku-4-5-20251001

25 - claude-opus-4-8

26 - claude-opus-4-7

27 - claude-opus-4-6

28 - claude-sonnet-4-6

29 - claude-opus-4-5-20251101

30 - claude-sonnet-4-5-20250929

1131---

1232

1333Cloud sandboxes run as isolated Linux containers on Anthropic-managed infrastructure. They come pre-installed with a comprehensive set of programming languages, databases, and utilities. The agent can use these immediately without any installation steps.

managed-agents/define-outcomes Changed · +62 / -0 lines

## Availability ### API support #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Required beta headers ### Data handling

from line 8

88 url: https://platform.claude.com/docs/en/managed-agents/overview

99 status: beta

1010 betaHeader: managed-agents-2026-04-01

11 zdr:

12 eligibility: not-eligible

13 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

14 supportedModels:

15 - claude-fable-5-1

16 - claude-mythos-5-1

17 - claude-fable-5

18 - claude-mythos-5

19 - claude-mythos-preview

20 - claude-opus-5-5

21 - claude-opus-5

22 - claude-sonnet-5-5

23 - claude-sonnet-5

24 - claude-haiku-5-5

25 - claude-haiku-4-5-20251001

26 - claude-opus-4-8

27 - claude-opus-4-7

28 - claude-opus-4-6

29 - claude-sonnet-4-6

30 - claude-opus-4-5-20251101

31 - claude-sonnet-4-5-20250929

32 supportedPlatforms:

33 Claude API: beta

34 Claude Platform on AWS: beta

35 Claude Models on Google Cloud (Anthropic managed): beta

1136---

1237

1338An outcome tells the session what the end result should look like and how to measure its quality. The agent works toward that target, self-evaluating and iterating until the outcome is met.

from line 892

867892 Upload files and mount them in your sandbox for reading and processing.

868893 </Card>

869894</CardGroup>

895

896## Availability

897

898Outcomes are in beta on every model.

899

900### API support

901

902#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

903

904* Availability: Available

905

906| Status | Cloud platforms |

907| ------ | ----------------------- |

908| Beta | All supported platforms |

909

910### Required beta headers

911

912[Beta headers overview](https://platform.claude.com/docs/en/api/beta-headers)

913

914Beta headers allow you to access experimental features and new model capabilities before they become part of the standard API.

915

916* `managed-agents-2026-04-01`

917

918### Data handling

919

920[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

921

922| Item | Zero data retention | HIPAA |

923| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |

924| Status | Not eligible | Not eligible |

925| Details | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes. | Applies to all Managed Agents sub-features, including Self-hosted sandboxes. |

926

927Applicable endpoints:

928

929* `/v1/agents`

930* `/v1/sessions`

931* `/v1/environments`

870932

managed-agents/dreams Changed · +62 / -3 lines

## Availability ### API support #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Required beta headers ### Data handling

from line 2

22title: Dreams

33url: https://platform.claude.com/docs/en/managed-agents/dreams

44description: Let Claude review past sessions to curate an agent's memory and surface new insights.

5featureMetadata:

6 status: research-preview

7 betaHeader: managed-agents-2026-04-01, dreaming-2026-04-21

8 zdr:

9 eligibility: not-eligible

10 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

11 supportedModels:

12 - claude-fable-5-1

13 - claude-fable-5

14 - claude-mythos-preview

15 - claude-opus-5-5

16 - claude-opus-5

17 - claude-sonnet-5-5

18 - claude-sonnet-5

19 - claude-opus-4-8

20 - claude-opus-4-7

21 - claude-sonnet-4-6

22 supportedPlatforms:

23 Claude API: research-preview

524---

625

7<Tip>

8 Dreaming is a research preview feature. [Request access](https://claude.com/form/claude-managed-agents) to try it.

9</Tip>

26<Note>

27 Dreams are in research preview.

1028

29 * [Request access](https://claude.com/form/claude-managed-agents)

30</Note>

1132Agents write to their [memory stores](https://platform.claude.com/docs/en/managed-agents/memory) as they work, but these writes are local and incremental: over many sessions a memory store accumulates duplicates, contradictions, and stale entries.

1233

1334**Dreams** let Claude clean that up. A dream reads an existing memory store alongside past session transcripts, then produces a new, reorganized memory store: duplicates merged, stale or contradicted entries replaced with the latest value, and new insights surfaced.

from line 656

635656| Supported models | `claude-opus-5-5`, `claude-fable-5-1`, `claude-opus-5`, `claude-fable-5`, `claude-opus-4-8`, `claude-opus-4-7`, `claude-sonnet-5-5`, `claude-sonnet-5`, `claude-sonnet-4-6` |

636657

637658Default rate limits apply to dream creation while this feature is in research preview. [Contact support](https://support.claude.com) if you need higher limits.

659

660## Availability

661

662Dreams are in research preview on select models.

663

664### API support

665

666#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

667

668* Availability: Available

669

670| Status | Cloud platforms | Models |

671| ---------------- | --------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

672| Research preview | Claude API | [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview), [Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/overview), Claude Mythos Preview, [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview), [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/overview), [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview), [Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/overview), [Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-4-8/overview), [Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-4-7/overview), [Claude Sonnet 4.6](https://platform.claude.com/docs/en/models/sonnet-4-6/overview) |

673

674### Required beta headers

675

676[Beta headers overview](https://platform.claude.com/docs/en/api/beta-headers)

677

678Beta headers allow you to access experimental features and new model capabilities before they become part of the standard API.

679

680* `managed-agents-2026-04-01`

681* `dreaming-2026-04-21`

682

683### Data handling

684

685[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

686

687| Item | Zero data retention | HIPAA |

688| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |

689| Status | Not eligible | Not eligible |

690| Details | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes. | Applies to all Managed Agents sub-features, including Self-hosted sandboxes. |

691

692Applicable endpoints:

693

694* `/v1/agents`

695* `/v1/sessions`

696* `/v1/environments`

638697

managed-agents/environments Changed · +58 / -1 lines

## Availability ### API support #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Required beta headers ### Data handling

from line 6

66 topic:

77 title: Managed Agents

88 url: https://platform.claude.com/docs/en/managed-agents/overview

9 status: beta

109 betaHeader: managed-agents-2026-04-01

10 zdr:

11 eligibility: not-eligible

12 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

13 supportedModels:

14 - claude-fable-5-1

15 - claude-mythos-5-1

16 - claude-fable-5

17 - claude-mythos-5

18 - claude-mythos-preview

19 - claude-opus-5-5

20 - claude-opus-5

21 - claude-sonnet-5-5

22 - claude-sonnet-5

23 - claude-haiku-5-5

24 - claude-haiku-4-5-20251001

25 - claude-opus-4-8

26 - claude-opus-4-7

27 - claude-opus-4-6

28 - claude-sonnet-4-6

29 - claude-opus-4-5-20251101

30 - claude-sonnet-4-5-20250929

1131---

1232

1333Environments define the sandbox configuration where your agent runs. You create an environment once, then reference its ID each time you start a session. Multiple sessions can share the same environment, but each session gets its own isolated sandbox (a fresh Linux container).

from line 820

800820 Create a session to run your agent and start running tasks.

801821 </Card>

802822</CardGroup>

823

824## Availability

825

826Cloud sandboxes are available on every model.

827

828### API support

829

830#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

831

832* Availability: Available

833

834| Status | Cloud platforms | Models |

835| --------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------- |

836| Available | Claude API, [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws), [Claude Models on Google Cloud (Anthropic managed)](https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/anthropic-managed) | [All models](https://platform.claude.com/docs/en/models/overview) |

837

838### Required beta headers

839

840[Beta headers overview](https://platform.claude.com/docs/en/api/beta-headers)

841

842Beta headers allow you to access experimental features and new model capabilities before they become part of the standard API.

843

844* `managed-agents-2026-04-01`

845

846### Data handling

847

848[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

849

850| Item | Zero data retention | HIPAA |

851| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |

852| Status | Not eligible | Not eligible |

853| Details | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes. | Applies to all Managed Agents sub-features, including Self-hosted sandboxes. |

854

855Applicable endpoints:

856

857* `/v1/agents`

858* `/v1/sessions`

859* `/v1/environments`

803860

managed-agents/event-deltas Changed · +62 / -0 lines

## Availability ### API support #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Required beta headers ### Data handling

from line 8

88 url: https://platform.claude.com/docs/en/managed-agents/overview

99 status: beta

1010 betaHeader: managed-agents-2026-04-01

11 zdr:

12 eligibility: not-eligible

13 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

14 supportedModels:

15 - claude-fable-5-1

16 - claude-mythos-5-1

17 - claude-fable-5

18 - claude-mythos-5

19 - claude-mythos-preview

20 - claude-opus-5-5

21 - claude-opus-5

22 - claude-sonnet-5-5

23 - claude-sonnet-5

24 - claude-haiku-5-5

25 - claude-haiku-4-5-20251001

26 - claude-opus-4-8

27 - claude-opus-4-7

28 - claude-opus-4-6

29 - claude-sonnet-4-6

30 - claude-opus-4-5-20251101

31 - claude-sonnet-4-5-20250929

32 supportedPlatforms:

33 Claude API: beta

34 Claude Platform on AWS: beta

35 Claude Models on Google Cloud (Anthropic managed): beta

1136---

1237

1338By default, the agent’s response text reaches the [session event stream](https://platform.claude.com/docs/en/managed-agents/events-and-streaming) as buffered `agent.message` events. Each one is emitted only after the model request that produced it completes. Event deltas let you render that text incrementally, as a live preview, while the model is still generating it.

from line 907

882907 Coordinate multiple agents within a single session.

883908 </Card>

884909</CardGroup>

910

911## Availability

912

913Event deltas are in beta on every model.

914

915### API support

916

917#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

918

919* Availability: Available

920

921| Status | Cloud platforms | Models |

922| ------ | ----------------------- | ----------------------------------------------------------------- |

923| Beta | All supported platforms | [All models](https://platform.claude.com/docs/en/models/overview) |

924

925### Required beta headers

926

927[Beta headers overview](https://platform.claude.com/docs/en/api/beta-headers)

928

929Beta headers allow you to access experimental features and new model capabilities before they become part of the standard API.

930

931* `managed-agents-2026-04-01`

932

933### Data handling

934

935[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

936

937| Item | Zero data retention | HIPAA |

938| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |

939| Status | Not eligible | Not eligible |

940| Details | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes. | Applies to all Managed Agents sub-features, including Self-hosted sandboxes. |

941

942Applicable endpoints:

943

944* `/v1/agents`

945* `/v1/sessions`

946* `/v1/environments`

885947

managed-agents/events-and-streaming Changed · +25 / -0 lines

from line 8

88 url: https://platform.claude.com/docs/en/managed-agents/overview

99 status: beta

1010 betaHeader: managed-agents-2026-04-01

11 zdr:

12 eligibility: not-eligible

13 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

14 supportedModels:

15 - claude-fable-5-1

16 - claude-mythos-5-1

17 - claude-fable-5

18 - claude-mythos-5

19 - claude-mythos-preview

20 - claude-opus-5-5

21 - claude-opus-5

22 - claude-sonnet-5-5

23 - claude-sonnet-5

24 - claude-haiku-5-5

25 - claude-haiku-4-5-20251001

26 - claude-opus-4-8

27 - claude-opus-4-7

28 - claude-opus-4-6

29 - claude-sonnet-4-6

30 - claude-opus-4-5-20251101

31 - claude-sonnet-4-5-20250929

32 supportedPlatforms:

33 Claude API: beta

34 Claude Platform on AWS: beta

35 Claude Models on Google Cloud (Anthropic managed): beta

1136---

1237

1338Communication with Claude Managed Agents is event-based. You open a stream to follow a session, send events to direct it, and answer it when it pauses for input.

managed-agents/files Changed · +25 / -0 lines

from line 8

88 url: https://platform.claude.com/docs/en/managed-agents/overview

99 status: beta

1010 betaHeader: managed-agents-2026-04-01

11 zdr:

12 eligibility: not-eligible

13 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

14 supportedModels:

15 - claude-fable-5-1

16 - claude-mythos-5-1

17 - claude-fable-5

18 - claude-mythos-5

19 - claude-mythos-preview

20 - claude-opus-5-5

21 - claude-opus-5

22 - claude-sonnet-5-5

23 - claude-sonnet-5

24 - claude-haiku-5-5

25 - claude-haiku-4-5-20251001

26 - claude-opus-4-8

27 - claude-opus-4-7

28 - claude-opus-4-6

29 - claude-sonnet-4-6

30 - claude-opus-4-5-20251101

31 - claude-sonnet-4-5-20250929

32 supportedPlatforms:

33 Claude API: beta

34 Claude Platform on AWS: beta

35 Claude Models on Google Cloud (Anthropic managed): beta

1136---

1237

1338You can provide files to your agent by uploading them through the Files API and mounting them in the session’s sandbox.

managed-agents/github Changed · +62 / -0 lines

## Availability ### API support #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Required beta headers ### Data handling

from line 8

88 url: https://platform.claude.com/docs/en/managed-agents/overview

99 status: beta

1010 betaHeader: managed-agents-2026-04-01

11 zdr:

12 eligibility: not-eligible

13 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

14 supportedModels:

15 - claude-fable-5-1

16 - claude-mythos-5-1

17 - claude-fable-5

18 - claude-mythos-5

19 - claude-mythos-preview

20 - claude-opus-5-5

21 - claude-opus-5

22 - claude-sonnet-5-5

23 - claude-sonnet-5

24 - claude-haiku-5-5

25 - claude-haiku-4-5-20251001

26 - claude-opus-4-8

27 - claude-opus-4-7

28 - claude-opus-4-6

29 - claude-sonnet-4-6

30 - claude-opus-4-5-20251101

31 - claude-sonnet-4-5-20250929

32 supportedPlatforms:

33 Claude API: beta

34 Claude Platform on AWS: beta

35 Claude Models on Google Cloud (Anthropic managed): beta

1136---

1237

1338You can mount a GitHub repository to your session sandbox and connect to the GitHub MCP for making pull requests.

from line 900

875900 Mount files in the sandbox alongside your repositories

876901 </Card>

877902</CardGroup>

903

904## Availability

905

906GitHub access is in beta on every model.

907

908### API support

909

910#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

911

912* Availability: Available

913

914| Status | Cloud platforms |

915| ------ | ----------------------- |

916| Beta | All supported platforms |

917

918### Required beta headers

919

920[Beta headers overview](https://platform.claude.com/docs/en/api/beta-headers)

921

922Beta headers allow you to access experimental features and new model capabilities before they become part of the standard API.

923

924* `managed-agents-2026-04-01`

925

926### Data handling

927

928[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

929

930| Item | Zero data retention | HIPAA |

931| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |

932| Status | Not eligible | Not eligible |

933| Details | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes. | Applies to all Managed Agents sub-features, including Self-hosted sandboxes. |

934

935Applicable endpoints:

936

937* `/v1/agents`

938* `/v1/sessions`

939* `/v1/environments`

878940

managed-agents/mcp-connector Changed · +25 / -0 lines

from line 8

88 url: https://platform.claude.com/docs/en/managed-agents/overview

99 status: beta

1010 betaHeader: managed-agents-2026-04-01

11 zdr:

12 eligibility: not-eligible

13 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

14 supportedModels:

15 - claude-fable-5-1

16 - claude-mythos-5-1

17 - claude-fable-5

18 - claude-mythos-5

19 - claude-mythos-preview

20 - claude-opus-5-5

21 - claude-opus-5

22 - claude-sonnet-5-5

23 - claude-sonnet-5

24 - claude-haiku-5-5

25 - claude-haiku-4-5-20251001

26 - claude-opus-4-8

27 - claude-opus-4-7

28 - claude-opus-4-6

29 - claude-sonnet-4-6

30 - claude-opus-4-5-20251101

31 - claude-sonnet-4-5-20250929

32 supportedPlatforms:

33 Claude API: beta

34 Claude Platform on AWS: beta

35 Claude Models on Google Cloud (Anthropic managed): beta

1136---

1237

1338Claude Managed Agents supports connecting [Model Context Protocol (MCP)](https://modelcontextprotocol.io) servers to your agents. This gives the agent access to external tools, data sources, and services through a standardized protocol.

managed-agents/memory Changed · +74 / -0 lines

## Availability ### API support #### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) ### Required beta headers ### Data handling

from line 5

55featureMetadata:

66 status: beta

77 betaHeader: agent-memory-2026-07-22

8 zdr:

9 eligibility: not-eligible

10 note: Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes.

11 supportedModels:

12 - claude-fable-5-1

13 - claude-mythos-5-1

14 - claude-fable-5

15 - claude-mythos-5

16 - claude-mythos-preview

17 - claude-opus-5-5

18 - claude-opus-5

19 - claude-sonnet-5-5

20 - claude-sonnet-5

21 - claude-haiku-5-5

22 - claude-haiku-4-5-20251001

23 - claude-opus-4-8

24 - claude-opus-4-7

25 - claude-opus-4-6

26 - claude-sonnet-4-6

27 - claude-opus-4-5-20251101

28 - claude-sonnet-4-5-20250929

29 supportedPlatforms:

30 Claude API:

31 availability: beta

32 note: Sending `managed-agents-2026-04-01` together with `agent-memory-2026-07-22` on a memory store call returns a 400 error.

33 Claude Platform on AWS:

34 availability: beta

35 note: On Claude Platform on AWS, sessions on self-hosted environments can't attach memory stores. Sending `managed-agents-2026-04-01` together with `agent-memory-2026-07-22` on a memory store call returns a 400 error.

36 Claude Models on Google Cloud (Anthropic managed):

37 availability: beta

38 note: On Claude Models on Google Cloud (Anthropic managed), memory stores aren't supported for sessions on self-hosted environments. Sending `managed-agents-2026-04-01` together with `agent-memory-2026-07-22` on a memory store call returns a 400 error.

839---

940

1041Each Managed Agents session starts with a fresh context by default. When a session ends, any state the agent built up is gone. Memory stores let the agent carry information across sessions: user preferences, project conventions, prior mistakes, and domain context.

from line 1368

13371368* **Attach a new store when it makes sense.** If a store has grown beyond its useful scope, attach a fresh one for new content and attach the original with `read_only` access. The agent can read from both while only writing to the new one.

13381369

13391370* **Limit write access where appropriate.** Sessions that only read shared reference material don’t need `read_write`. Keeping write access scoped to sessions that actually add new memories makes it easier to track where growth is coming from.

1371

1372## Availability

1373

1374Memory stores are in beta on every model.

1375

1376### API support

1377

1378#### [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)

1379

1380* Availability: Available

1381

1382| Status | Cloud platforms |

1383| ---------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- |

1384| Beta1 | Claude API |

1385| Beta2 | [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) |

1386| Beta3 | [Claude Models on Google Cloud (Anthropic managed)](https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/anthropic-managed) |

1387

13881. Sending `managed-agents-2026-04-01` together with `agent-memory-2026-07-22` on a memory store call returns a 400 error.

13892. On Claude Platform on AWS, sessions on self-hosted environments can’t attach memory stores. Sending `managed-agents-2026-04-01` together with `agent-memory-2026-07-22` on a memory store call returns a 400 error.

13903. On Claude Models on Google Cloud (Anthropic managed), memory stores aren’t supported for sessions on self-hosted environments. Sending `managed-agents-2026-04-01` together with `agent-memory-2026-07-22` on a memory store call returns a 400 error.

1391

1392### Required beta headers

1393

1394[Beta headers overview](https://platform.claude.com/docs/en/api/beta-headers)

1395

1396Beta headers allow you to access experimental features and new model capabilities before they become part of the standard API.

1397

1398* `agent-memory-2026-07-22`

1399

1400### Data handling

1401

1402[Data handling overview](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility)

1403

1404| Item | Zero data retention | HIPAA |

1405| ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |

1406| Status | Not eligible | Not eligible |

1407| Details | Sessions are stateful resources; transcripts persist until you delete them. Applies to all Managed Agents sub-features, including Self-hosted sandboxes. | Applies to all Managed Agents sub-features, including Self-hosted sandboxes. |

1408

1409Applicable endpoints:

1410

1411* `/v1/agents`

1412* `/v1/sessions`

1413* `/v1/environments`

13401414