One read of Claude Developer Platformapi-20261005T233724Z
17 pages moved out of 753 read.
Pages moved
17
significant first
Pages read
753
in this capture
Captured
23:37 UTC
Corpus hash
dcae3122f81a
corpus-hash
What this read moved
1-17 of 17api/beta Changed · +30 / -16 lines
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
Nothing in the body moved in this read. What changed is above.
api/beta/organization Changed · +28 / -14 lines
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
Nothing in the body moved in this read. What changed is above.
api/beta/organization/rbac_roles Changed · +41 / -21 lines
from line 49
4949
5050 format: date-time
5151
52 - `name: string`
52 - `display_name: string`
5353
54 Name of the RBAC Role.
54 Name of the RBAC Role. For a role created by Anthropic, this name can differ from the label claude.ai shows, and Anthropic may change the name. To keep a lasting reference to a role, store its `id`.
5555
5656 - `updated_at: string`
5757
from line 59
5959
6060 format: date-time
6161
62 - `name: string`
63
64 **Deprecated**: Use `display_name` instead; `name` always has the same value.
65
66 Deprecated: use `display_name` instead. Name of the RBAC Role; always the same value as `display_name`.
67
6268- `has_more: boolean`
6369
6470 Indicates whether there are more results beyond this page.
from line 90
8490 {
8591 "id": "rbac_role_016J8xVtKpDq3Wy9ZmN2hR4s",
8692 "created_at": "2024-10-30T23:58:27.427722Z",
93 "display_name": "Project Editor",
8794 "name": "Project Editor",
8895 "type": "rbac_role",
8996 "updated_at": "2024-10-30T23:58:27.427722Z"
from line 137
130137
131138 format: date-time
132139
133 - `name: string`
140 - `display_name: string`
134141
135 Name of the RBAC Role.
142 Name of the RBAC Role. For a role created by Anthropic, this name can differ from the label claude.ai shows, and Anthropic may change the name. To keep a lasting reference to a role, store its `id`.
136143
137144 - `updated_at: string`
138145
from line 147
140147
141148 format: date-time
142149
150 - `name: string`
151
152 **Deprecated**: Use `display_name` instead; `name` always has the same value.
153
154 Deprecated: use `display_name` instead. Name of the RBAC Role; always the same value as `display_name`.
155
143156### Example
144157
145158```bash
from line 167
154167{
155168 "id": "rbac_role_016J8xVtKpDq3Wy9ZmN2hR4s",
156169 "created_at": "2024-10-30T23:58:27.427722Z",
170 "display_name": "Project Editor",
157171 "name": "Project Editor",
158172 "type": "rbac_role",
159173 "updated_at": "2024-10-30T23:58:27.427722Z"
from line 198
184198
185199 format: date-time
186200
187 - `name: string`
201 - `display_name: string`
188202
189 Name of the RBAC Role.
203 Name of the RBAC Role. For a role created by Anthropic, this name can differ from the label claude.ai shows, and Anthropic may change the name. To keep a lasting reference to a role, store its `id`.
190204
191205 - `updated_at: string`
192206
from line 207
193207 RFC 3339 datetime string indicating when the RBAC Role was last updated.
194208
195209 format: date-time
210
211 - `name: string`
212
213 **Deprecated**: Use `display_name` instead; `name` always has the same value.
214
215 Deprecated: use `display_name` instead. Name of the RBAC Role; always the same value as `display_name`.
196216
197217## RBAC Roles › Permissions
198218
api/beta/organization/rbac_roles/list Changed · +9 / -2 lines
from line 47
4747
4848 format: date-time
4949
50 - `name: string`
50 - `display_name: string`
5151
52 Name of the RBAC Role.
52 Name of the RBAC Role. For a role created by Anthropic, this name can differ from the label claude.ai shows, and Anthropic may change the name. To keep a lasting reference to a role, store its `id`.
5353
5454 - `updated_at: string`
5555
from line 57
5757
5858 format: date-time
5959
60 - `name: string`
61
62 **Deprecated**: Use `display_name` instead; `name` always has the same value.
63
64 Deprecated: use `display_name` instead. Name of the RBAC Role; always the same value as `display_name`.
65
6066- `has_more: boolean`
6167
6268 Indicates whether there are more results beyond this page.
from line 88
8288 {
8389 "id": "rbac_role_016J8xVtKpDq3Wy9ZmN2hR4s",
8490 "created_at": "2024-10-30T23:58:27.427722Z",
91 "display_name": "Project Editor",
8592 "name": "Project Editor",
8693 "type": "rbac_role",
8794 "updated_at": "2024-10-30T23:58:27.427722Z"
api/beta/organization/rbac_roles/retrieve Changed · +9 / -2 lines
from line 39
3939
4040 format: date-time
4141
42 - `name: string`
42 - `display_name: string`
4343
44 Name of the RBAC Role.
44 Name of the RBAC Role. For a role created by Anthropic, this name can differ from the label claude.ai shows, and Anthropic may change the name. To keep a lasting reference to a role, store its `id`.
4545
4646 - `updated_at: string`
4747
from line 49
4949
5050 format: date-time
5151
52 - `name: string`
53
54 **Deprecated**: Use `display_name` instead; `name` always has the same value.
55
56 Deprecated: use `display_name` instead. Name of the RBAC Role; always the same value as `display_name`.
57
5258## Example
5359
5460```bash
from line 69
6369{
6470 "id": "rbac_role_016J8xVtKpDq3Wy9ZmN2hR4s",
6571 "created_at": "2024-10-30T23:58:27.427722Z",
72 "display_name": "Project Editor",
6673 "name": "Project Editor",
6774 "type": "rbac_role",
6875 "updated_at": "2024-10-30T23:58:27.427722Z"
manage-claude/inference-hooks Changed · +5 / -1 lines
from line 12
1212
1313Because the hook runs on Anthropic's servers, after the request leaves the client and before the model runs, it applies to every governed request uniformly, with nothing to install or deploy on user devices.
1414
15Today the only hook event is `prompt`, which fires once per governed inference request, before inference begins. Response-side enforcement is planned as a later event.
15There are two hook events. `prompt` fires once per governed inference request, before inference begins. `tool_call` fires when Claude's response contains tool calls, before any of them runs, in organizations that have turned on **Validate tool calls**.
1616
1717***
1818
from line 27
2727
2828
2929
30With **Validate tool calls** on, Claude's tool calls are a third hooked point, which the diagram doesn't show: your AI security server returns one verdict for all the tool calls in a response before any of them runs.
31
3032A verdict is a small JSON object: `{"action": "allow"}` lets the request proceed, and a deny carries the user-facing reason. For the full verdict schema, see [Return a verdict](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#return-a-verdict).
3133
3234Your AI security server sees what the user sees: transcript text, tool calls and their results, and text extracted from attachments. It never receives raw file or image bytes, system prompts, or Anthropic-internal context.
from line 80
7880One hook governs conversations across claude.ai, Cowork, Claude Code, and Claude Tag sessions in your Claude Enterprise organization, whether they run on the web, in the desktop or mobile apps, in the CLI, or in Slack. Inference hooks are not available on Amazon Bedrock or Google Cloud.
7981
8082Governed requests are the inference requests behind the user's conversation. Ancillary requests, such as conversation title generation, aren't sent to your endpoint, and system prompts and tool definitions are never included in what is sent. Voice mode is not covered.
83
84With **Validate tool calls** on, `tool_call` events may leave out calls to some of claude.ai's own tools, such as tools that suggest connectors or skills, give Claude the user's local time, or read the user's own past chats and saved memory. These calls and their results stay in the conversation, so the next `prompt` event from that conversation, if there is one, includes them.
8185
8286Some features that Anthropic runs for your organization make model calls of their own. Your organization sees the results of those calls but not their transcripts. These calls aren't governed requests and aren't sent to your endpoint. They include the following:
8387
manage-claude/inference-hooks-endpoint Changed · +115 / -21 lines
## The tool call frame ### Platform tools ### Application tools ### Third-party tools ### Client tools
from line 158
158158| `User-Agent` | `anthropic-dlp/1` |
159159| `Accept-Encoding` | `identity` |
160160
161There is one hook event today: the prompt frame, sent once per governed inference request, before inference begins. Anthropic holds the request until your AI security server responds or the verdict timeout elapses.
161There are two hook events, told apart by the top-level `type` field. The prompt frame is sent once per governed inference request, before inference begins. The tool call frame is sent when a model response contains tool calls, before any of them runs, in organizations that have turned on **Validate tool calls**. Either way, Anthropic waits until your AI security server responds or the verdict timeout elapses.
162162
163163## The prompt frame
164164
165165The request body is a JSON object with these fields:
166166
167| Field | Type | Description |
168| ------------ | -------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
169| `type` | string | The hook event. Always `"prompt"` today; other event types will be introduced in the future, so handle an unrecognized value gracefully (see [Forward compatibility](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#forward-compatibility)). |
170| `request_id` | string | Opaque per-inference-call identifier for correlation. Equals the `webhook-id` header. |
171| `tenant_id` | string or null | Opaque identifier for the organization the request belongs to. |
172| `actor` | object | The principal the request is attributed to, discriminated on `type` (`"user"` is the only value sent today): `id` (a tagged identifier, stable across requests for the same account) and `email_address` (when available). Both `id` and `email_address` can be null. |
173| `source` | object | The originating application: `application` (see [Source values](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#source-values)). |
174| `messages` | array | The conversation transcript up to the point of inference. See [Content blocks](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#content-blocks). |
175| `session_id` | string or null | Opaque conversation identifier, when one exists. Don't parse it. For Claude Code it is a best-effort, client-asserted session identifier. |
176| `model` | string or null | Public model identifier for this request, when available. |
177| `metadata` | object | Reserved extension map of string keys to string values, sent empty today. Require nothing from it, and tolerate its absence, its presence, and any keys that appear. |
167| Field | Type | Description |
168| ------------ | -------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
169| `type` | string | The hook event: `"prompt"` or `"tool_call"` (see [The tool call frame](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#the-tool-call-frame)). Other event types will be introduced in the future, so handle an unrecognized value gracefully (see [Forward compatibility](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#forward-compatibility)). |
170| `request_id` | string | Opaque per-frame identifier for correlation. Equals the `webhook-id` header. |
171| `tenant_id` | string or null | Opaque identifier for the organization the request belongs to. |
172| `actor` | object | The principal the request is attributed to, discriminated on `type` (`"user"` is the only value sent today): `id` (a tagged identifier, stable across requests for the same account) and `email_address` (when available). Both `id` and `email_address` can be null. |
173| `source` | object | The originating application: `application` (see [Source values](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#source-values)). |
174| `messages` | array | The conversation transcript up to the point of inference. See [Content blocks](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#content-blocks). |
175| `session_id` | string or null | Opaque conversation identifier, when one exists. Don't parse it. For Claude Code it is a best-effort, client-asserted session identifier. |
176| `model` | string or null | Public model identifier for this request, when available. |
177| `metadata` | object | Reserved extension map of string keys to string values, sent empty today. Require nothing from it, and tolerate its absence, its presence, and any keys that appear. |
178178
179179An example request body:
180180
from line 219
219219
220220Each entry in `messages` has a `role` of `user` or `assistant` (tool results appear under the `user` role, matching the public Messages API content model) and a `content` array of blocks discriminated by `type`:
221221
222| Block `type` | Fields |
223| ------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
224| `text` | `text`: the text content. |
225| `tool_use` | `id`: the identifier the matching tool result references. `tool_name`: the tool's name. `input`: the arguments the model passed to the tool. |
226| `tool_result` | `content`: the tool's output as text, with parts joined by newlines; binary parts such as images are replaced by placeholder markers, and raw bytes are never sent. `is_error`: whether the tool call failed. `tool_name`: the tool's name, so a policy can condition on tool identity without cross-referencing an earlier block. `tool_use_id`: the `id` of the matching `tool_use` block. |
227| `attachment` | `file_name`: the original file name or path. `media_type`: the attachment's media type. `size_bytes`: the size of the original file. `text`: the text content of the attachment when available, such as extracted document text, an audio transcript, or link metadata. Raw attachment bytes are never sent. |
222| Block `type` | Fields |
223| ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
224| `text` | `text`: the text content. |
225| `tool_use` | `id`: the identifier the matching tool result references. `tool_name`: the tool's name. `input`: the arguments the model passed to the tool. `tool_info`: on a tool call frame only (left out elsewhere, never `null`), an object that says who provides the tool; see [The tool call frame](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#the-tool-call-frame). |
226| `tool_result` | `content`: the tool's output as text, with parts joined by newlines; binary parts such as images are replaced by placeholder markers, and raw bytes are never sent. `is_error`: whether the tool call failed. `tool_name`: the tool's name, so a policy can condition on tool identity without cross-referencing an earlier block. `tool_use_id`: the `id` of the matching `tool_use` block. |
227| `attachment` | `file_name`: the original file name or path. `media_type`: the attachment's media type. `size_bytes`: the size of the original file. `text`: the text content of the attachment when available, such as extracted document text, an audio transcript, or link metadata. Raw attachment bytes are never sent. |
228228
229229Apart from `type`, a `text` block's `text`, and a `tool_result` block's `content` and `is_error`, any of these fields can be `null` when the value isn't known; for example, an image arrives as an `attachment` block with `file_name` and `text` set to `null`.
230230
from line 244
244244
245245Treat `source.application` as advisory routing metadata, not a trust boundary: don't rest a security-critical policy decision on it alone.
246246
247## The tool call frame
248
249When Claude produces tool calls, Anthropic sends one tool call frame that lists them. Calls to some of claude.ai's own tools may be left out, and a response whose only tool calls are left out doesn't produce a tool call frame; see [Availability](https://platform.claude.com/docs/en/manage-claude/inference-hooks#availability). One verdict covers the whole frame: you can't allow some tool calls and deny others. The frame goes to the same endpoint as the prompt frame, with the same headers, signature, and top-level fields. Tool calls made by code that Claude runs in the code execution tool are sent the same way, in separate tool call frames, before they run.
250
251It differs from the prompt frame in three ways:
252
253* `type` is `"tool_call"`.
254* `messages` holds only the latest message, the `assistant` message Claude just produced: any `text` blocks and one `tool_use` block per tool call the frame lists, in the order the model produced them. Earlier conversation is left out, because the prompt frame sent before that model call carried it. Read the last entry of `messages`, because the protocol may later add earlier messages before it.
255* Each `tool_use` block carries a `tool_info` object that says who provides the tool.
256
257Where `session_id` is set, it is the same on both frames. The tool call frame has its own `request_id`, which is opaque like the prompt frame's.
258
259`tool_info` says who provides the tool, not who runs it or what it can reach. It is one of four kinds, told apart by its `type` field, and each kind carries its own fields. Anthropic sends the first of the following kinds that fits the tool. New kinds may appear: accept a `type` you don't recognize, and for such a kind rely only on `type`.
260
261An optional field that doesn't apply is left out, never `null`, so a `tool_info` can be just `{"type": "client"}`. `tool_name` is chosen by whoever defined the tool, and a server's `toolset_name` by whoever wrote the request, so don't treat either as a trust boundary.
262
263### Platform tools
264
265A platform tool is one of the Claude API's predefined tools, such as web search or bash. It is `"platform"` whether Anthropic or the application that calls Claude runs it.
266
267| Field | Present | Description |
268| -------------- | -------- | ------------------------------------------------------------------------------------------------------------------- |
269| `type` | Always | `"platform"` |
270| `tool_type` | Always | The tool's versioned type, such as `web_search_20250305`. Match it exactly; don't parse a name or a date out of it. |
271| `toolset_name` | Optional | The toolset the tool belongs to, such as `browser`. |
272
273### Application tools
274
275An application tool is one that the Anthropic application making the request provides itself, such as claude.ai's own tools. It is `"application"` whether Anthropic or the application that calls Claude runs it.
276
277| Field | Present | Description |
278| -------------- | -------- | --------------------------------- |
279| `type` | Always | `"application"` |
280| `toolset_name` | Optional | The group of tools it belongs to. |
281
282### Third-party tools
283
284A third-party tool is one on a server that Anthropic knows of, such as a claude.ai connector or an MCP server that the request names in `mcp_servers`. This doesn't mean Anthropic has vetted the server.
285
286| Field | Present | Description |
287| ----------------- | -------------------- | ---------------------------------------------------------------------------------------------------------------------------------- |
288| `type` | Always | `"third_party"` |
289| `toolset_name` | Optional | The name the request gives the server. |
290| `origin` | Optional | The scheme, host, and non-default port of the server's URL. An absent `origin` means unknown, not safe. |
291| `verified_origin` | Whenever `origin` is | `true` only when Anthropic's servers connect to the server themselves. Treat `false` as an origin that Anthropic hasn't confirmed. |
292
293A `tool_use` block for a tool on an MCP server that the request names `crm` in `mcp_servers`:
294
295```json
296{
297 "type": "tool_use",
298 "id": "toolu_01GhIjKlMnOpQrStUvWxYzAb",
299 "tool_name": "crm_search",
300 "input": {
301 "query": "accounts renewing in Q4"
302 },
303 "tool_info": {
304 "type": "third_party",
305 "toolset_name": "crm",
306 "origin": "https://mcp.crm.example.com",
307 "verified_origin": true
308 }
309}
310```
311
312### Client tools
313
314A client tool is any other tool. The application that calls Claude declares it and receives its calls. A call to a tool the request doesn't declare is also `"client"`.
315
316| Field | Present | Description |
317| -------------- | -------- | --------------------------------- |
318| `type` | Always | `"client"` |
319| `toolset_name` | Optional | The group of tools it belongs to. |
320
321A `tool_use` block for a tool that the calling application declares:
322
323```json
324{
325 "type": "tool_use",
326 "id": "toolu_01AbCdEfGhIjKlMnOpQrStUv",
327 "tool_name": "read_file",
328 "input": {
329 "path": "reports/q3.txt"
330 },
331 "tool_info": {
332 "type": "client"
333 }
334}
335```
336
247337## Return a verdict
248338
249339Respond with HTTP 200 and a JSON verdict body for both outcomes; the `action` field discriminates. To allow the request:
from line 369
279369
280370Anthropic reads at most 64 KiB of the response body, and the body must be uncompressed. Redirects are not followed, and cookies are ignored. Unknown fields in the verdict body are ignored, so you can return a richer object alongside the fields documented here.
281371
372A tool call frame takes the same verdict body, and every rule in this section applies to it unchanged. `allow` lets the tool calls run and the response continues; `deny` stops all of them and ends the response with the same error as a denied prompt, including your `deny_reason`. Text delivered before the first tool call is not taken back.
373
282374## Verify the signature
283375
284376Requests are signed per the [Standard Webhooks](https://www.standardwebhooks.com/) specification, using three headers. Anthropic sends the header names in lowercase, and proxies are free to re-case them, so look them up case-insensitively.
from line 767
675767
676768### Timeout and retry
677769
678Your administrator sets a verdict timeout between 1 and 10,000ms (5,000ms by default). The budget covers the entire exchange: connection, TLS handshake, request, and response.
770Your administrator sets a verdict timeout between 1 and 10,000ms (5,000ms by default). The budget covers the entire exchange: connection, TLS handshake, request, and response. The same timeout applies to tool call frames.
679771
680772Anthropic retries exactly once, after a 100ms delay, and only when the connection attempt fails. The retry shares the same timeout budget and carries the same `webhook-id` and the same signature. Once your AI security server has responded, the exchange is never retried.
681773
from line 781
689781
690782Starting 10 minutes after the trip, Anthropic checks whether your server has recovered: at most about once per minute it sends your server the same synthetic test request that **Test connection** sends (`source.application` is `config-test`), signed like any other request and carrying no user content. Respond to it normally. A valid verdict, allow or deny, resets the breaker and enforcement resumes; a webhook failure leaves the breaker tripped, and the checks continue. An administrator can also reset the breaker at any time, and administrator configuration changes stop the automatic checks; see [Circuit breaker](https://platform.claude.com/docs/en/manage-claude/inference-hooks-configuration#circuit-breaker).
691783
784Prompt frames and tool call frames share one circuit breaker, and failures on either count toward it.
785
692786Each trip is recorded as an `inference_hooks_circuit_breaker_tripped` activity in the [Activity Feed](https://platform.claude.com/docs/en/manage-claude/compliance-activity-feed), one activity per trip. While the breaker is tripped, no per-request Inference hooks activities are recorded, so the trip activity is the feed's only record of the tripped window.
693787
694788### Latency
695789
696Enforcement adds your AI security server's round trip to the latency of every governed request in your organization. Keep the verdict fast, and load-test your server before rolling it out to a large organization.
790Enforcement adds your AI security server's round trip to the latency of every governed request in your organization. With **Validate tool calls** on, a response that produces a tool call frame also waits for its verdict, in shadow mode as well. Keep the verdict fast, and load-test your server before rolling it out to a large organization.
697791
698792### Source IP addresses
699793
from line 797
703797
704798The protocol grows without breaking correctly written servers. Your server must ignore:
705799
706* Unknown top-level fields on the prompt frame.
800* Unknown top-level fields on either frame.
707801* Unknown keys in `metadata`.
708802* New `source.application` values.
709803* New `actor.type` values. `actor` is a union discriminated on `type`, and `"user"` is the only kind sent today; a future kind guarantees only that `type` is present.
models/fable-5/introducing-claude-fable-5-and-claude-mythos-5 Changed · +6 / -6 lines
from line 14
1414
1515Claude Fable 5 is built for demanding reasoning and long-horizon agentic work. Claude Mythos 5 shares the same capabilities and is available only in limited release through [Project Glasswing](https://anthropic.com/glasswing).
1616
17The headline change for integrations: Claude Fable 5 includes safety classifiers that can decline requests. Claude Mythos 5 does not include these classifiers. If your integration calls Claude Fable 5, plan for three changes: new response handling for refusals, fallback options for retrying on another Claude model, and new billing rules. [Refusals, fallback, and billing on Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/introducing-claude-fable-5-and-claude-mythos-5#refusals-fallback-and-billing-on-claude-fable-5) summarizes all three.
17The headline change for integrations: Claude Fable 5 includes safety classifiers that can decline requests. If your integration calls Claude Fable 5, plan for three changes: new response handling for refusals, fallback options for retrying on another Claude model, and new billing rules. [Refusals, fallback, and billing on Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/introducing-claude-fable-5-and-claude-mythos-5#refusals-fallback-and-billing-on-claude-fable-5) summarizes all three.
1818
1919## Models
2020
21| Model | API model ID | Description |
22| --------------- | ----------------- | --------------------------------------------------------------------------------------------------------------------------------------------- |
23| Claude Fable 5 | `claude-fable-5` | Built for demanding reasoning and long-horizon agentic work |
24| Claude Mythos 5 | `claude-mythos-5` | Shares Claude Fable 5's capabilities without the safety classifiers. Available through Project Glasswing. Successor to Claude Mythos Preview. |
21| Model | API model ID | Description |
22| --------------- | ----------------- | -------------------------------------------------------------------------------------------------------------- |
23| Claude Fable 5 | `claude-fable-5` | Built for demanding reasoning and long-horizon agentic work |
24| Claude Mythos 5 | `claude-mythos-5` | Shares Claude Fable 5's capabilities. Available through Project Glasswing. Successor to Claude Mythos Preview. |
2525
2626Claude Fable 5 and Claude Mythos 5 share the same specs and pricing:
2727
from line 32
3232
3333## Refusals, fallback, and billing on Claude Fable 5
3434
35Claude Fable 5 includes safety classifiers that can decline certain requests. Claude Mythos 5 does not include these classifiers, so this section applies to Claude Fable 5 only. The following sections summarize what refusals mean for your integration. Each links to the full guide.
35Claude Fable 5 includes safety classifiers that can decline certain requests. The following sections summarize what refusals mean for your integration. Each links to the full guide.
3636
3737### Refusals
3838
models/fable-5/migration-guide Changed · +4 / -5 lines
from line 31
3131Where the two models diverge:
3232
3333* **Availability:** Claude Fable 5 does not require access approval. Claude Mythos 5 is available only to approved customers in [Project Glasswing](https://anthropic.com/glasswing).
34* **Safety classifiers:** Claude Fable 5 runs safety classifiers that can decline requests with `stop_reason: "refusal"`. Claude Mythos 5 does not include these classifiers. See [Refusals and fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback).
3534* **Priority Tier:** [Priority Tier](https://platform.claude.com/docs/en/api/service-tiers#supported-models) is supported on Claude Fable 5 but not on Claude Mythos 5.
3635
3736## Migrating to Claude Mythos 5 and Claude Fable 5 from Claude Mythos Preview
from line 37
3837
3938[Claude Mythos 5](https://anthropic.com/glasswing) is the access-gated successor to [Claude Mythos Preview](https://anthropic.com/glasswing), the invitation-only research preview. [Claude Fable 5](https://platform.claude.com/docs/en/models/fable-5/introducing-claude-fable-5-and-claude-mythos-5) offers the same capabilities and does not require access approval. The changes in this section apply equally to both targets.
4039
41Migration is mostly drop-in. Claude Mythos 5 and Claude Fable 5 use the same [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) and the same [tool use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview) patterns as Claude Mythos Preview, and token counts are roughly unchanged because all three models use the same tokenizer. The key changes to check are the features that are no longer available (listed in the next section) and thinking output. If you migrate to Claude Fable 5, also plan for safety classifier refusals, which Claude Mythos Preview and Claude Mythos 5 do not have; see [Refusals and fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback).
40Migration is mostly drop-in. Claude Mythos 5 and Claude Fable 5 use the same [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) and the same [tool use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview) patterns as Claude Mythos Preview, and token counts are roughly unchanged because all three models use the same tokenizer. The key changes to check are the features that are no longer available (listed in the next section) and thinking output. If you migrate to Claude Fable 5, also plan for safety classifier refusals; see [Refusals and fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback).
4241
4342For the Claude Mythos Preview retirement timeline, see [Model deprecations](https://platform.claude.com/docs/en/about-claude/model-deprecations).
4443
from line 322
323322* Remove `budget_tokens`. It has no direct replacement: thinking is adaptive, and the `effort` parameter is a separate output-level control, not a thinking budget.
324323* Verify any code that parses the `thinking` field treats it as display text only and passes thinking blocks back unchanged when continuing on the same model. `thinking.display` defaults to `"omitted"` on `claude-mythos-5` and `claude-fable-5`, the same as on Claude Mythos Preview. Set `display: "summarized"` to receive readable summaries. See [Thinking output on Claude Fable and Claude Mythos models](https://platform.claude.com/docs/en/build-with-claude/thinking#thinking-output-on-claude-fable-5-and-claude-mythos-5).
325324* If you replay conversation history on an earlier model, strip `thinking` and `redacted_thinking` blocks from prior assistant turns first. Thinking blocks from `claude-fable-5` and `claude-mythos-5` are readable only by the model that produced them or a newer one: earlier models silently ignore them, while Claude Fable 5.1 and Claude Mythos 5.1 read them, so keep them when you move a conversation up to those models (see [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)). Stripping keeps requests to earlier models minimal and uniform.
326* If you migrate to Claude Fable 5, handle `stop_reason: "refusal"` and read the `stop_details.category` field. Claude Fable 5 runs safety classifiers that Claude Mythos Preview and Claude Mythos 5 do not have. See [Refusals and fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback).
325* If you migrate to Claude Fable 5, handle `stop_reason: "refusal"` and read the `stop_details.category` field. See [Refusals and fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback).
327326* Re-baseline token counts and costs on your own workloads. Token counts are roughly unchanged when migrating from `claude-mythos-preview`.
328327
329328## Migrating to Claude Mythos 5 and Claude Fable 5 from Claude Opus 5
from line 365
366365 If your code is on Claude Opus 4.7 or earlier, first apply the relevant [Migrating to Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide) from-section for the API-level changes from your current model, then the remaining delta in this section.
367366</Note>
368367
369Migration is mostly drop-in. Claude Fable 5 and Claude Mythos 5 use the same [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) and the same [tool use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview) patterns as Claude Opus 4.8, with the same [1M token context window](https://platform.claude.com/docs/en/build-with-claude/context-windows) by default and the same [128k max output tokens](https://platform.claude.com/docs/en/models/overview). Token counts are roughly unchanged because the models use the same tokenizer. The key changes to check are always-on [adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking), thinking output, safety classifier refusals (Claude Fable 5 only), and pricing.
368Migration is mostly drop-in. Claude Fable 5 and Claude Mythos 5 use the same [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) and the same [tool use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview) patterns as Claude Opus 4.8, with the same [1M token context window](https://platform.claude.com/docs/en/build-with-claude/context-windows) by default and the same [128k max output tokens](https://platform.claude.com/docs/en/models/overview). Token counts are roughly unchanged because the models use the same tokenizer. The key changes to check are always-on [adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking), thinking output, safety classifier refusals, and pricing.
370369
371370### Update your model name
372371
from line 677
678677
6796784. **Thinking output:** On `claude-fable-5` and `claude-mythos-5`, the raw chain of thought is never returned, but thinking blocks still carry readable summarized text when `thinking.display` is set to `summarized`. Pass thinking blocks back unchanged when continuing a conversation on the same model. See [Thinking output on Claude Fable and Claude Mythos models](https://platform.claude.com/docs/en/build-with-claude/thinking#thinking-output-on-claude-fable-5-and-claude-mythos-5).
680679
6815. **Safety classifiers and the `refusal` stop reason (Claude Fable 5 only):** `claude-fable-5` runs safety classifiers on requests and during response generation. Claude Mythos 5 does not include these classifiers. When a classifier declines a request, the Messages API returns `stop_reason: "refusal"` as a successful HTTP 200 response, not an error. The `stop_details.category` field reports which classifier fired, with categories such as `"cyber"`, `"bio"`, and `"reasoning_extraction"`, or `null` when the refusal maps to no named category. See the [refusal category table](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#refusal-response) for the full set.
6805. **Safety classifiers and the `refusal` stop reason:** `claude-fable-5` runs safety classifiers on requests and during response generation. When a classifier declines a request, the Messages API returns `stop_reason: "refusal"` as a successful HTTP 200 response, not an error. The `stop_details.category` field reports which classifier fired, with categories such as `"cyber"`, `"bio"`, and `"reasoning_extraction"`, or `null` when the refusal maps to no named category. See the [refusal category table](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#refusal-response) for the full set.
682681
683682 A refusal that arrives before any output is billed when its category is `"bio"`, `"frontier_llm"`, or `"reasoning_extraction"`. A refusal before any output in any other category, or with a `null` category, is not billed ([How refusals are billed](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#how-refusals-are-billed)). Before September 24, 2026, no refusal before any output was billed on Claude Fable 5. When a classifier fires mid-stream, the input and already-streamed output are billed; discard the partial output.
684683
api/beta/agents Changed · +2 / -2 lines
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
from line 413
413413
414414- `tools: optional array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams`
415415
416 Tool configurations available to the agent. Maximum of 128 tools across all toolsets allowed.
416 Tool configurations available to the agent. Maximum of 256 tools across all toolsets allowed.
417417
418418 - `BetaManagedAgentsAgentToolset20260401Params object`
419419
from line 3586
35863586
35873587- `tools: optional array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams or null`
35883588
3589 Tool configurations available to the agent. Full replacement. Omit to preserve; send empty array or null to clear. Maximum of 128 tools across all toolsets allowed.
3589 Tool configurations available to the agent. Full replacement. Omit to preserve; send empty array or null to clear. Maximum of 256 tools across all toolsets allowed.
35903590
35913591 - `BetaManagedAgentsAgentToolset20260401Params object`
35923592
api/beta/agents/create Changed · +1 / -1 lines
from line 411
411411
412412- `tools: optional array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams`
413413
414 Tool configurations available to the agent. Maximum of 128 tools across all toolsets allowed.
414 Tool configurations available to the agent. Maximum of 256 tools across all toolsets allowed.
415415
416416 - `BetaManagedAgentsAgentToolset20260401Params object`
417417
api/beta/agents/update Changed · +1 / -1 lines
from line 417
417417
418418- `tools: optional array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams or null`
419419
420 Tool configurations available to the agent. Full replacement. Omit to preserve; send empty array or null to clear. Maximum of 128 tools across all toolsets allowed.
420 Tool configurations available to the agent. Full replacement. Omit to preserve; send empty array or null to clear. Maximum of 256 tools across all toolsets allowed.
421421
422422 - `BetaManagedAgentsAgentToolset20260401Params object`
423423
manage-claude/inference-hooks-configuration Changed · +3 / -1 lines
from line 88
8888 To evaluate verdicts against live traffic without blocking anyone at first, set **Mode** to **Shadow mode** (step 6) before turning on enforcement; see [Shadow mode](https://platform.claude.com/docs/en/manage-claude/inference-hooks-configuration#shadow-mode).
8989
9090 Turn on **Enforce verdicts** to gate Claude on your AI security server's verdict for every governed prompt, then confirm in the dialog, which restates your failure handling choice. Allow about a minute for the change to reach every Anthropic server; requests already in flight finish under the old setting. Turning it off stops prompts from being sent to your AI security server, again within about a minute; your configuration is kept.
91
92 Turn on **Validate tool calls**, below **Enforce verdicts**, and confirm in the dialog, to also send the tool calls in each of Claude's responses to your AI security server and wait for its verdict before they run; see [The tool call frame](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#the-tool-call-frame). It is off by default. It has no effect while **Enforce verdicts** is off, and a change to it takes about a minute to reach every Anthropic server, as with **Enforce verdicts**. Confirm that your AI security server handles tool call frames before you turn it on.
9193 </Step>
9294</Steps>
9395
9496## Shadow mode
9597
96Shadow mode runs your hook against live traffic without blocking anything. Your AI security server receives governed prompts and returns verdicts exactly as it would when enforcing, but nothing is blocked: every request proceeds to the model, even when your server denies it or can't be reached, and the end user sees nothing. Use it to tune your policy against your organization's real traffic before you start enforcing.
98Shadow mode runs your hook against live traffic without blocking anything. Your AI security server receives governed prompts and returns verdicts exactly as it would when enforcing, but nothing is blocked: every request proceeds to the model, even when your server denies it or can't be reached, and the end user sees nothing. Use it to tune your policy against your organization's real traffic before you start enforcing. With **Validate tool calls** on, your server also receives tool call frames in shadow mode, and no tool call is blocked.
9799
98100To use shadow mode, set **Mode** to **Shadow mode** under **Failure handling**, then turn on **Enforce verdicts** so prompts flow to your AI security server. While it is active, the settings page shows a **Shadow mode — not blocking** badge. To leave shadow mode, set **Mode** back to **Allow the request** or **Block the request**; verdicts are enforced again once enforcement is on.
99101
models/fable-5-1/migration-guide Changed · +0 / -1 lines
from line 33
3333Where the two models diverge:
3434
3535* **Availability:** Claude Fable 5.1 doesn't require access approval. Claude Mythos 5.1 is available only to approved customers in [Project Glasswing](https://anthropic.com/glasswing). Contact your Anthropic account team for access.
36* **Safety classifiers:** Claude Fable 5.1 runs safety classifiers covering the same `stop_details` categories as Claude Fable 5. A declined request returns `stop_reason: "refusal"` with a `stop_details.category`, and can fall back to another model with the `fallbacks` parameter or a client-side retry. See [Refusals and fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback).
3736* **Priority Tier:** Neither model is supported on [Priority Tier](https://platform.claude.com/docs/en/api/service-tiers#supported-models). Claude Fable 5 is.
3837
3938## Migrating to Claude Fable 5.1 from Claude Fable 5
models/fable-5-1/overview Changed · +1 / -1 lines
from line 16
1616
1717## Overview
1818
19Claude Fable 5.1 extends Claude Fable 5 at the same input and output prices, with cache reads at a quarter of the cost, and brings stronger long-running agentic coding, multistep research, and document, spreadsheet, and slide work. For most workloads, start with Claude Opus 5 (see [Choosing a model](https://platform.claude.com/docs/en/about-claude/models/choosing-a-model)). Use Claude Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your evals on Claude Opus 5 at higher effort still fall short. Claude Mythos 5.1 offers the same capabilities to [Project Glasswing](https://anthropic.com/glasswing) participants only.
19Claude Fable 5.1 extends Claude Fable 5 at the same input and output prices, with cache reads at a quarter of the cost, and brings stronger long-running agentic coding, multistep research, and document, spreadsheet, and slide work. For most workloads, start with Claude Opus 5.5 (see [Choosing a model](https://platform.claude.com/docs/en/about-claude/models/choosing-a-model)). Use Claude Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your evals on Claude Opus 5.5 at higher effort still fall short. Claude Mythos 5.1 offers the same capabilities to [Project Glasswing](https://anthropic.com/glasswing) participants only.
2020
2121If you already call Claude Fable 5, three changes are breaking: [forced tool use returns an error](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#forced-tool-use-is-not-supported), [earlier models can't read its thinking blocks](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#thinking-blocks-are-tied-to-the-model-that-produced-them), and [editing earlier turns invalidates thinking blocks](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#editing-earlier-turns-invalidates-thinking-blocks). Five are additive: [per-message effort](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#change-effort-mid-conversation-beta) (beta), [turn-scoped system messages](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#turn-scoped-system-messages-beta) (beta), [readable progress updates between tool calls](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#progress-updates-between-tool-calls-beta) (`display: "updates"`, beta), a [lower cache read price](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#pricing), and [content provenance](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#content-provenance).
2222
models/fable-5-1/whats-new-fable-5-1 Changed · +1 / -1 lines
from line 4
44description: Overview of new features, breaking changes, and capability improvements in Claude Fable 5.1 and Claude Mythos 5.1.
55---
66
7Claude Fable 5.1 extends Claude Fable 5 at the same input and output prices, with cache reads at a quarter of the cost, and brings stronger long-running agentic coding, multistep research, and document, spreadsheet, and slide work. For most workloads, start with Claude Opus 5 (see [Choosing a model](https://platform.claude.com/docs/en/about-claude/models/choosing-a-model)). Use Claude Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your evals on Claude Opus 5 at higher effort still fall short. Claude Mythos 5.1 offers the same capabilities to [Project Glasswing](https://anthropic.com/glasswing) participants only.
7Claude Fable 5.1 extends Claude Fable 5 at the same input and output prices, with cache reads at a quarter of the cost, and brings stronger long-running agentic coding, multistep research, and document, spreadsheet, and slide work. For most workloads, start with Claude Opus 5.5 (see [Choosing a model](https://platform.claude.com/docs/en/about-claude/models/choosing-a-model)). Use Claude Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your evals on Claude Opus 5.5 at higher effort still fall short. Claude Mythos 5.1 offers the same capabilities to [Project Glasswing](https://anthropic.com/glasswing) participants only.
88
99If you already call Claude Fable 5, three changes are breaking: [forced tool use returns an error](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#forced-tool-use-is-not-supported), [earlier models can't read its thinking blocks](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#thinking-blocks-are-tied-to-the-model-that-produced-them), and [editing earlier turns invalidates thinking blocks](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#editing-earlier-turns-invalidates-thinking-blocks). Five are additive: [per-message effort](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#change-effort-mid-conversation-beta) (beta), [turn-scoped system messages](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#turn-scoped-system-messages-beta) (beta), [readable progress updates between tool calls](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#progress-updates-between-tool-calls-beta) (`display: "updates"`, beta), a [lower cache read price](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#pricing), and [content provenance](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1#content-provenance).
1010
models/fable-5/overview Changed · +1 / -1 lines
from line 16
1616
1717## Fable vs. Mythos
1818
19[Claude Mythos 5](https://platform.claude.com/docs/en/models/mythos-5/overview) is offered separately, by invitation only, for defensive cybersecurity workflows as part of [Project Glasswing](https://anthropic.com/glasswing). It shares Claude Fable 5's specifications and pricing; Claude Fable 5 includes safety classifiers that can decline requests, and Claude Mythos 5 does not. For access, contact your Anthropic, AWS, or Google Cloud account team.
19[Claude Mythos 5](https://platform.claude.com/docs/en/models/mythos-5/overview) is offered separately, by invitation only, for defensive cybersecurity workflows as part of [Project Glasswing](https://anthropic.com/glasswing). It shares Claude Fable 5's specifications and pricing. For access, contact your Anthropic, AWS, or Google Cloud account team.
2020
2121## How it compares to the current lineup
2222