One read of Claude Developer Platformapi-20261008T020712Z

This capture is too large to show at once. Changes 126-136 of 136 are below, significant first; the rest are on the following screens.

api/organization/federation/rules/create Changed · +1 / -1 lines

from line 206

206206 

207207 - `target: ServiceAccountTarget`

208208 

209 Identity that tokens minted via this rule act as. Currently always a `service_account` target.

209 What this rule targets. Check `type` before reading the other fields. Tokens minted via a rule whose target `type` is `service_account` act as that service account.

210210 

211211 - `type: "service_account"`

212212 

api/organization/federation/rules/list Changed · +1 / -1 lines

from line 124

124124 

125125 - `target: ServiceAccountTarget`

126126 

127 Identity that tokens minted via this rule act as. Currently always a `service_account` target.

127 What this rule targets. Check `type` before reading the other fields. Tokens minted via a rule whose target `type` is `service_account` act as that service account.

128128 

129129 - `type: "service_account"`

130130 

api/organization/federation/rules/retrieve Changed · +1 / -1 lines

from line 116

116116 

117117 - `target: ServiceAccountTarget`

118118 

119 Identity that tokens minted via this rule act as. Currently always a `service_account` target.

119 What this rule targets. Check `type` before reading the other fields. Tokens minted via a rule whose target `type` is `service_account` act as that service account.

120120 

121121 - `type: "service_account"`

122122 

api/organization/federation/rules/update Changed · +1 / -1 lines

from line 210

210210 

211211 - `target: ServiceAccountTarget`

212212 

213 Identity that tokens minted via this rule act as. Currently always a `service_account` target.

213 What this rule targets. Check `type` before reading the other fields. Tokens minted via a rule whose target `type` is `service_account` act as that service account.

214214 

215215 - `type: "service_account"`

216216 

build-with-claude/claude-on-amazon-bedrock-legacy Changed · +0 / -1 lines

from line 153

153153| Claude Sonnet 4.5 ([deprecated](https://platform.claude.com/docs/en/about-claude/model-deprecations)) | `anthropic.claude-sonnet-4-5-20250929-v1:0` | Yes | Yes | Yes | Yes | No |

154154| Claude Sonnet 4 ([deprecated](https://platform.claude.com/docs/en/about-claude/model-deprecations)) | `anthropic.claude-sonnet-4-20250514-v1:0` | Yes | Yes | Yes | No | Yes |

155155| Claude Haiku 4.5 | `anthropic.claude-haiku-4-5-20251001-v1:0` | Yes | Yes | Yes | No | No |

156| Claude Haiku 3.5 ([deprecated](https://platform.claude.com/docs/en/about-claude/model-deprecations)) | `anthropic.claude-3-5-haiku-20241022-v1:0` | No | Yes | No | No | No |

157156 

158157### List available models

159158 

build-with-claude/preserved-thinking Changed · +1 / -1 lines

from line 41

4141Claude Sonnet 5.5 reads thinking blocks from Claude Sonnet 5, Claude Opus 4.8, Claude Haiku 4.5, and earlier models, and, on the Claude API and Google Cloud, from Claude Haiku 5.5, but not from Claude Opus 5, Claude Opus 5.5, or any Claude Fable or Claude Mythos model. On the Claude API and Google Cloud, Claude Opus 5.5 reads thinking blocks from Claude Sonnet 5.5; no other model does. So a conversation that moves from Claude Sonnet 5 onto Claude Sonnet 5.5 keeps its reasoning, and so does one that moves from Claude Sonnet 5.5 up to Claude Opus 5.5 on the Claude API and Google Cloud. One that moves onto Claude Sonnet 5.5 from Claude Opus 5, Claude Opus 5.5, or a Claude Fable or Claude Mythos model runs the turns after the switch without the previous model's reasoning. So does any other move away from Claude Sonnet 5.5, for example a [server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback) to Claude Sonnet 5.

4242 

4343* **A conversation that moves to Claude Fable 5.1 from an earlier model, or from Claude Opus 5.5 on the Claude API, keeps its reasoning.** The earlier model's thinking blocks stay readable, so the model thinks as usual from the first turn after the switch.

44* **A conversation that moves down to an earlier model loses Claude Fable 5.1's reasoning for that request.** This happens when a router sends a turn to a cheaper model, after a [classifier refusal fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback), or during a [server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback). The API removes the unreadable blocks before the prompt reaches the model. They aren't billed and don't count toward `input_tokens`.

44* **A conversation that moves down to an earlier model loses Claude Fable 5.1's reasoning for that request.** This occurs when a router sends a turn to a cheaper model, after a [classifier refusal fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback), or during a [server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback). The API removes the unreadable blocks before the prompt reaches the model. They aren't billed and don't count toward `input_tokens`.

4545 

4646Keep sending the full history on every request, thinking blocks included, and let the API drop what the current model can't read. The API never edits your `messages` array, so the dropped blocks stay in your history. When the same history goes back to Claude Fable 5.1, its blocks are readable again, along with the earlier model's thinking. The reasoning is lost for good only if your client removes the blocks itself, for example a harness that strips thinking on a model switch or rebuilds the history from what each model used.

4747 

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

from line 377

377377 

378378The `session.usage` event is a snapshot of the session's cumulative usage and tracked list cost. It carries the session's token totals, `list_cost`, `active_seconds`, `server_tool_use` request counts (`web_search_requests`, priced into list cost per request, and `web_fetch_requests`, which reads `0` because web fetch requests carry no per-request charge and aren't metered), and an echo of the session's `budget`, or `null` when the session has none. It appears in the events list and the session stream. The session emits one immediately before it goes idle, whatever the stop reason, so a session that reaches its budget always emits one immediately before the budget-reached idle event.

379379 

380To read usage from the stream and the session object, see [Tracking usage](https://platform.claude.com/docs/en/managed-agents/events-and-streaming#tracking-usage).

380To read usage from the stream and the session object, see [Track usage](https://platform.claude.com/docs/en/managed-agents/session-observability#track-usage).

381381 

382382## Budgets in multiagent sessions

383383 

managed-agents/mcp-connector Changed · +1 / -1 lines

from line 386

386386 

387387### Handle connection and authentication failures

388388 

389Session creation does not validate MCP connectivity or credentials. It does check each declared server's host against the environment's networking: with a `limited` environment, session creation fails with a 400 error when a host is not allowed, as described under [Provide authentication at session creation](https://platform.claude.com/docs/en/managed-agents/mcp-connector#provide-authentication-at-session-creation). If an MCP server is unreachable or rejects the supplied credential, the session still starts and interaction remains possible. A [`session.error`](https://platform.claude.com/docs/en/managed-agents/events-and-streaming) event is emitted with the `mcp_server_name` of the affected server and a `retry_status`:

389Session creation does not validate MCP connectivity or credentials. It does check each declared server's host against the environment's networking: with a `limited` environment, session creation fails with a 400 error when a host is not allowed, as described under [Provide authentication at session creation](https://platform.claude.com/docs/en/managed-agents/mcp-connector#provide-authentication-at-session-creation). If an MCP server is unreachable or rejects the supplied credential, the session still starts and interaction remains possible. A [`session.error`](https://platform.claude.com/docs/en/managed-agents/reference#event-types) event is emitted with the `mcp_server_name` of the affected server and a `retry_status`:

390390 

391391| Error type | Meaning |

392392| --------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |

managed-agents/multiagent-orchestration Changed · +1 / -1 lines

from line 1189

11891189 

11901190Critical events are proxied to the primary thread. However, you might still want to investigate a specific agent's reasoning and tool calls. To do so, stream or list the events from the associated session thread.

11911191 

1192Each session thread has its own event stream at `/v1/sessions/{session_id}/threads/{thread_id}/stream`, and it accepts the same `event_deltas[]` parameter as the session-level stream, so you can preview a subagent's text as the model generates it. A connection previews only the thread it's reading: a child thread's previews never appear on the session-level stream, so to watch a subagent live, open its own thread stream. See [Preview session thread events](https://platform.claude.com/docs/en/managed-agents/events-and-streaming#preview-session-thread-events) for opting in, accumulating, and reconciling previews.

1192Each session thread has its own event stream at `/v1/sessions/{session_id}/threads/{thread_id}/stream`, and it accepts the same `event_deltas[]` parameter as the session-level stream, so you can preview a subagent's text as the model generates it. A connection previews only the thread it's reading: a child thread's previews never appear on the session-level stream, so to watch a subagent live, open its own thread stream. See [Preview session thread events](https://platform.claude.com/docs/en/managed-agents/event-deltas#preview-session-thread-events) for opting in, accumulating, and reconciling previews.

11931193 

11941194<Tabs>

11951195 <Tab title="Stream session thread events">

managed-agents/self-hosted-sandboxes-custom-tools Changed · +1 / -1 lines

from line 194

194194 </Step>

195195</Steps>

196196 

197The worker answers only the tools registered with it. If a tool is declared on the agent but no worker or client serves it, the session pauses with a `requires_action` stop reason. It stays paused until something posts the result. See [Handling custom tool calls](https://platform.claude.com/docs/en/managed-agents/events-and-streaming#handling-custom-tool-calls) for the event flow.

197The worker answers only the tools registered with it. If a tool is declared on the agent but no worker or client serves it, the session pauses with a `requires_action` stop reason. It stays paused until something posts the result. See [Answer tool calls that pause the session](https://platform.claude.com/docs/en/managed-agents/events-and-streaming#answer-tool-calls-that-pause-the-session) for the event flow.

198198 

199199## Wrap an MCP server as custom tools

200200 

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

from line 443

443443 

444444No other event type is accepted. Events that respond to an agent turn (`user.tool_confirmation`, `user.tool_result`, and `user.custom_tool_result`) aren't accepted because no agent turn exists yet, and `user.interrupt` isn't accepted because there is no turn to stop. Unlike `initial_events` on a scheduled deployment, a session's `initial_events` don't accept `system.message`.

445445 

446Each event in `initial_events` is validated and persisted before the create response returns, in list order, with a server-assigned ID, exactly as if you had posted it to the [send events](https://platform.claude.com/docs/en/managed-agents/events-and-streaming) endpoint immediately after creation. Per-event content rules are also the same as on that endpoint. An empty list is equivalent to omitting the field. Validation is all-or-nothing: if any event fails validation, the whole request is rejected and no session is created.

446Each event in `initial_events` is validated and persisted before the create response returns, in list order, with a server-assigned ID, exactly as if you had posted it to the [send events](https://platform.claude.com/docs/en/managed-agents/events-and-streaming#send-events) endpoint immediately after creation. Per-event content rules are also the same as on that endpoint. An empty list is equivalent to omitting the field. Validation is all-or-nothing: if any event fails validation, the whole request is rejected and no session is created.

447447 

448448The create request is rejected in the following cases:

449449