Sessions changedapi/beta/sessions
Nearest release: v2.1.258, published an hour after this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+253added
Lines−67removed
From line
14
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits31to this page, all time
The whole hunk
from line 14, old and new numbered
/
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 14
1414
1515 - `string`
1616
17 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 38 more`
17 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 41 more`
1818
1919 - `"message-batches-2024-09-24"`
2020
from line 98
9898
9999 - `"ce-user-management-2026-07-13"`
100100
101 - `"mid-conversation-output-config-2026-07-01"`
102
103 - `"thinking-binding-controls-2026-08-01"`
104
105 - `"mid-conversation-system-clear-at-2026-08-21"`
106
101107### Body parameters
102108
103109- `agent: string or BetaManagedAgentsAgentParams or BetaManagedAgentsAgentWithOverridesParams`
from line 164
158164
159165 Replacement model. Accepts the model string, e.g. `claude-opus-5`, or a `model_config` object. Omit to use the agent's model.
160166
161 - `BetaManagedAgentsModel = "claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more or string`
167 - `BetaManagedAgentsModel = "claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more or string`
162168
163169 The model that will power your agent.
164170
165171 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
166172
167 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
173 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
168174
169175 The model that will power your agent.
170176
171177 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
172178
179 - `"claude-fable-5-1"`
180
181 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
182
173183 - `"claude-sonnet-5"`
174184
175185 High-performance model for coding and agents
from line 1108
10981108
10991109 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
11001110
1101 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
1111 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
11021112
11031113 The model that will power your agent.
11041114
11051115 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
11061116
1117 - `"claude-fable-5-1"`
1118
1119 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
1120
11071121 - `"claude-sonnet-5"`
11081122
11091123 High-performance model for coding and agents
from line 1666
16521666
16531667 - `outcome_evaluations: array of BetaManagedAgentsOutcomeEvaluationResource`
16541668
1655 Per-outcome evaluation state. One entry per define_outcome event sent to the session.
1669 Per-outcome evaluation state. One entry per `define_outcome` event sent to the session.
16561670
16571671 - `completed_at: string or null`
16581672
from line 1680
16661680
16671681 - `explanation: string or null`
16681682
1669 Grader's verdict text from the most recent evaluation. For satisfied, explains why criteria are met; for needs_revision (intermediate), what's missing; for failed, why unrecoverable.
1683 Grader's verdict text from the most recent evaluation. For `satisfied`, explains why criteria are met; for `needs_revision` (intermediate), what's missing; for `failed`, why unrecoverable.
16701684
16711685 - `iteration: number`
16721686
from line 1808
17941808
17951809 - `active_seconds: optional number`
17961810
1797 Cumulative time in seconds the session spent in running status. Excludes idle time.
1811 Cumulative time in seconds the session spent in `running` status. Excludes idle time.
17981812
17991813 format: double
18001814
from line 2135
21212135
21222136- `agent_version: optional number`
21232137
2124 Filter by agent version. Only applies when agent_id is also set.
2138 Filter by agent version. Only applies when `agent_id` is also set.
21252139
21262140 format: int32
21272141
from line 2179
21652179
21662180- `memory_store_id: optional string`
21672181
2168 Filter sessions whose resources contain a memory_store with this memory store ID.
2182 Filter sessions whose resources contain a `memory_store` with this memory store ID.
21692183
21702184- `order: optional "asc" or "desc"`
21712185
2172 Sort direction for results, ordered by created_at. Defaults to desc (newest first).
2186 Sort direction for results, ordered by `created_at`. Defaults to `desc` (newest first).
21732187
21742188 - `"asc"`
21752189
from line 2213
21992213
22002214 - `string`
22012215
2202 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 38 more`
2216 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 41 more`
22032217
22042218 - `"message-batches-2024-09-24"`
22052219
from line 2297
22832297
22842298 - `"ce-user-management-2026-07-13"`
22852299
2300 - `"mid-conversation-output-config-2026-07-01"`
2301
2302 - `"thinking-binding-controls-2026-08-01"`
2303
2304 - `"mid-conversation-system-clear-at-2026-08-21"`
2305
22862306### Returns
22872307
22882308- `data: optional array of BetaManagedAgentsSession`
from line 2337
23172337
23182338 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
23192339
2320 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
2340 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
23212341
23222342 The model that will power your agent.
23232343
23242344 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
23252345
2346 - `"claude-fable-5-1"`
2347
2348 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
2349
23262350 - `"claude-sonnet-5"`
23272351
23282352 High-performance model for coding and agents
from line 2895
28712895
28722896 - `outcome_evaluations: array of BetaManagedAgentsOutcomeEvaluationResource`
28732897
2874 Per-outcome evaluation state. One entry per define_outcome event sent to the session.
2898 Per-outcome evaluation state. One entry per `define_outcome` event sent to the session.
28752899
28762900 - `completed_at: string or null`
28772901
from line 2909
28852909
28862910 - `explanation: string or null`
28872911
2888 Grader's verdict text from the most recent evaluation. For satisfied, explains why criteria are met; for needs_revision (intermediate), what's missing; for failed, why unrecoverable.
2912 Grader's verdict text from the most recent evaluation. For `satisfied`, explains why criteria are met; for `needs_revision` (intermediate), what's missing; for `failed`, why unrecoverable.
28892913
28902914 - `iteration: number`
28912915
from line 3037
30133037
30143038 - `active_seconds: optional number`
30153039
3016 Cumulative time in seconds the session spent in running status. Excludes idle time.
3040 Cumulative time in seconds the session spent in `running` status. Excludes idle time.
30173041
30183042 format: double
30193043
from line 3376
33523376
33533377 - `string`
33543378
3355 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 38 more`
3379 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 41 more`
33563380
33573381 - `"message-batches-2024-09-24"`
33583382
from line 3460
34363460
34373461 - `"ce-user-management-2026-07-13"`
34383462
3463 - `"mid-conversation-output-config-2026-07-01"`
3464
3465 - `"thinking-binding-controls-2026-08-01"`
3466
3467 - `"mid-conversation-system-clear-at-2026-08-21"`
3468
34393469### Returns
34403470
34413471- `BetaManagedAgentsSession object`
from line 3500
34703500
34713501 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
34723502
3473 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
3503 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
34743504
34753505 The model that will power your agent.
34763506
34773507 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
34783508
3509 - `"claude-fable-5-1"`
3510
3511 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
3512
34793513 - `"claude-sonnet-5"`
34803514
34813515 High-performance model for coding and agents
from line 4058
40244058
40254059 - `outcome_evaluations: array of BetaManagedAgentsOutcomeEvaluationResource`
40264060
4027 Per-outcome evaluation state. One entry per define_outcome event sent to the session.
4061 Per-outcome evaluation state. One entry per `define_outcome` event sent to the session.
40284062
40294063 - `completed_at: string or null`
40304064
from line 4072
40384072
40394073 - `explanation: string or null`
40404074
4041 Grader's verdict text from the most recent evaluation. For satisfied, explains why criteria are met; for needs_revision (intermediate), what's missing; for failed, why unrecoverable.
4075 Grader's verdict text from the most recent evaluation. For `satisfied`, explains why criteria are met; for `needs_revision` (intermediate), what's missing; for `failed`, why unrecoverable.
40424076
40434077 - `iteration: number`
40444078
from line 4200
41664200
41674201 - `active_seconds: optional number`
41684202
4169 Cumulative time in seconds the session spent in running status. Excludes idle time.
4203 Cumulative time in seconds the session spent in `running` status. Excludes idle time.
41704204
41714205 format: double
41724206
from line 4525
44914525
44924526 - `string`
44934527
4494 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 38 more`
4528 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 41 more`
44954529
44964530 - `"message-batches-2024-09-24"`
44974531
from line 4609
45754609
45764610 - `"ce-user-management-2026-07-13"`
45774611
4612 - `"mid-conversation-output-config-2026-07-01"`
4613
4614 - `"thinking-binding-controls-2026-08-01"`
4615
4616 - `"mid-conversation-system-clear-at-2026-08-21"`
4617
45784618### Body parameters
45794619
45804620- `agent: optional BetaManagedAgentsSessionAgentUpdate`
from line 5089
50495089
50505090 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
50515091
5052 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
5092 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
50535093
50545094 The model that will power your agent.
50555095
50565096 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
50575097
5098 - `"claude-fable-5-1"`
5099
5100 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
5101
50585102 - `"claude-sonnet-5"`
50595103
50605104 High-performance model for coding and agents
from line 5647
56035647
56045648 - `outcome_evaluations: array of BetaManagedAgentsOutcomeEvaluationResource`
56055649
5606 Per-outcome evaluation state. One entry per define_outcome event sent to the session.
5650 Per-outcome evaluation state. One entry per `define_outcome` event sent to the session.
56075651
56085652 - `completed_at: string or null`
56095653
from line 5661
56175661
56185662 - `explanation: string or null`
56195663
5620 Grader's verdict text from the most recent evaluation. For satisfied, explains why criteria are met; for needs_revision (intermediate), what's missing; for failed, why unrecoverable.
5664 Grader's verdict text from the most recent evaluation. For `satisfied`, explains why criteria are met; for `needs_revision` (intermediate), what's missing; for `failed`, why unrecoverable.
56215665
56225666 - `iteration: number`
56235667
from line 5789
57455789
57465790 - `active_seconds: optional number`
57475791
5748 Cumulative time in seconds the session spent in running status. Excludes idle time.
5792 Cumulative time in seconds the session spent in `running` status. Excludes idle time.
57495793
57505794 format: double
57515795
from line 6118
60746118
60756119 - `string`
60766120
6077 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 38 more`
6121 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 41 more`
60786122
60796123 - `"message-batches-2024-09-24"`
60806124
from line 6202
61586202
61596203 - `"ce-user-management-2026-07-13"`
61606204
6205 - `"mid-conversation-output-config-2026-07-01"`
6206
6207 - `"thinking-binding-controls-2026-08-01"`
6208
6209 - `"mid-conversation-system-clear-at-2026-08-21"`
6210
61616211### Returns
61626212
61636213- `BetaManagedAgentsDeletedSession object`
from line 6255
62056255
62066256 - `string`
62076257
6208 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 38 more`
6258 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 41 more`
62096259
62106260 - `"message-batches-2024-09-24"`
62116261
from line 6339
62896339
62906340 - `"ce-user-management-2026-07-13"`
62916341
6342 - `"mid-conversation-output-config-2026-07-01"`
6343
6344 - `"thinking-binding-controls-2026-08-01"`
6345
6346 - `"mid-conversation-system-clear-at-2026-08-21"`
6347
62926348### Returns
62936349
62946350- `BetaManagedAgentsSession object`
from line 6379
63236379
63246380 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
63256381
6326 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
6382 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
63276383
63286384 The model that will power your agent.
63296385
63306386 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
63316387
6388 - `"claude-fable-5-1"`
6389
6390 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
6391
63326392 - `"claude-sonnet-5"`
63336393
63346394 High-performance model for coding and agents
from line 6937
68776937
68786938 - `outcome_evaluations: array of BetaManagedAgentsOutcomeEvaluationResource`
68796939
6880 Per-outcome evaluation state. One entry per define_outcome event sent to the session.
6940 Per-outcome evaluation state. One entry per `define_outcome` event sent to the session.
68816941
68826942 - `completed_at: string or null`
68836943
from line 6951
68916951
68926952 - `explanation: string or null`
68936953
6894 Grader's verdict text from the most recent evaluation. For satisfied, explains why criteria are met; for needs_revision (intermediate), what's missing; for failed, why unrecoverable.
6954 Grader's verdict text from the most recent evaluation. For `satisfied`, explains why criteria are met; for `needs_revision` (intermediate), what's missing; for `failed`, why unrecoverable.
68956955
68966956 - `iteration: number`
68976957
from line 7079
70197079
70207080 - `active_seconds: optional number`
70217081
7022 Cumulative time in seconds the session spent in running status. Excludes idle time.
7082 Cumulative time in seconds the session spent in `running` status. Excludes idle time.
70237083
70247084 format: double
70257085
from line 7479
74197479
74207480 Replacement model. Accepts the model string, e.g. `claude-opus-5`, or a `model_config` object. Omit to use the agent's model.
74217481
7422 - `BetaManagedAgentsModel = "claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more or string`
7482 - `BetaManagedAgentsModel = "claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more or string`
74237483
74247484 The model that will power your agent.
74257485
74267486 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
74277487
7428 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
7488 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
74297489
74307490 The model that will power your agent.
74317491
74327492 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
74337493
7494 - `"claude-fable-5-1"`
7495
7496 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
7497
74347498 - `"claude-sonnet-5"`
74357499
74367500 High-performance model for coding and agents
from line 8425
83618425
83628426- `BetaManagedAgentsOutcomeEvaluationResource object`
83638427
8364 Evaluation state for a single outcome defined via a define_outcome event.
8428 Evaluation state for a single outcome defined via a `define_outcome` event.
83658429
83668430 - `completed_at: string or null`
83678431
from line 8439
83758439
83768440 - `explanation: string or null`
83778441
8378 Grader's verdict text from the most recent evaluation. For satisfied, explains why criteria are met; for needs_revision (intermediate), what's missing; for failed, why unrecoverable.
8442 Grader's verdict text from the most recent evaluation. For `satisfied`, explains why criteria are met; for `needs_revision` (intermediate), what's missing; for `failed`, why unrecoverable.
83798443
83808444 - `iteration: number`
83818445
from line 8509
84458509
84468510 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
84478511
8448 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
8512 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
84498513
84508514 The model that will power your agent.
84518515
84528516 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
84538517
8518 - `"claude-fable-5-1"`
8519
8520 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
8521
84548522 - `"claude-sonnet-5"`
84558523
84568524 High-performance model for coding and agents
from line 9067
89999067
90009068 - `outcome_evaluations: array of BetaManagedAgentsOutcomeEvaluationResource`
90019069
9002 Per-outcome evaluation state. One entry per define_outcome event sent to the session.
9070 Per-outcome evaluation state. One entry per `define_outcome` event sent to the session.
90039071
90049072 - `completed_at: string or null`
90059073
from line 9081
90139081
90149082 - `explanation: string or null`
90159083
9016 Grader's verdict text from the most recent evaluation. For satisfied, explains why criteria are met; for needs_revision (intermediate), what's missing; for failed, why unrecoverable.
9084 Grader's verdict text from the most recent evaluation. For `satisfied`, explains why criteria are met; for `needs_revision` (intermediate), what's missing; for `failed`, why unrecoverable.
90179085
90189086 - `iteration: number`
90199087
from line 9209
91419209
91429210 - `active_seconds: optional number`
91439211
9144 Cumulative time in seconds the session spent in running status. Excludes idle time.
9212 Cumulative time in seconds the session spent in `running` status. Excludes idle time.
91459213
91469214 format: double
91479215
from line 9341
92739341
92749342 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
92759343
9276 - `"claude-sonnet-5" or "claude-fable-5" or "claude-opus-5" or 10 more`
9344 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`
92779345
92789346 The model that will power your agent.
92799347
92809348 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
92819349
9350 - `"claude-fable-5-1"`
9351
9352 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
9353
92829354 - `"claude-sonnet-5"`
92839355
92849356 High-performance model for coding and agents
from line 9395
93239395
93249396 Powerful intelligence for long-running agents and coding
93259397
9326 - `"claude-sonnet-4-5"`
9327
9328 High-performance model for agents and coding
9329
9330 - `"claude-sonnet-4-5-20250929"`
9331
9332 High-performance model for agents and coding
9333
9334 - `string`
9335
9336 - `effort: optional BetaManagedAgentsEffortLow or BetaManagedAgentsEffortMedium or BetaManagedAgentsEffortHigh or 2 more`
9337
9338 How hard Claude works on each turn. Sets `output_config.effort` on every Messages call the session makes.
9339
9340 - `BetaManagedAgentsEffortLow object`
9341
9342 Low effort. Favors latency over reasoning depth.
9343
9344 - `type: "low"`
9345
9346 - `BetaManagedAgentsEffortMedium object`
9347
9348 Medium effort. Balances latency and reasoning depth.
9349
9350 - `type: "medium"`
9351
9352 - `BetaManagedAgentsEffortHigh object`
9353
9354 High effort. Favors reasoning depth.
9355
9356 - `type: "high"`
9357
9358 - `BetaManagedAgentsEffortXhigh object`
9359
9360 Extra-high effort. Not all models accept this level.
9361
9362 - `type: "xhigh"`
9363
9364 - `BetaManagedAgentsEffortMax object`
9365
9366 Maximum effort. Favors reasoning depth over latency.
9367
9368 - `type: "max"`
9369
9370 - `inference_geo: optional string`
9371
9372 Geographic region for model inference. When unset, requests fall through to the workspace's default_inference_geo.
9373
9374 - `speed: optional "standard" or "fast"`
9375
9376 Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time.
9377
9378 - `"standard"`
9379
9380 - `"fast"`
9381
9382 - `multiagent: BetaManagedAgentsSessionMultiagentCoordinator or null`
9383
9384 Resolved coordinator topology with full agent definitions for each roster member.
9385
9386 - `agents: array of BetaManagedAgentsSessionThreadAgent or BetaManagedAgentsAdvisor`
9387
9388 Full `agent` definitions the coordinator may spawn as session threads.
9389
9390 - `BetaManagedAgentsSessionThreadAgent object`
9391
9392 Resolved `agent` definition for a single `session_thread`. Snapshot of the agent at thread creation time. The multiagent roster is not repeated here; read it from `Session.agent`.
9393
9394 - `id: string`
9395
9396 - `description: string or null`
9397
9398 - `mcp_servers: array of BetaManagedAgentsMCPServerURLDefinition`
9399
9400 - `name: string`
9401
9402 - `type: "url"`
9403
9404 - `url: string`
9405
9406 - `model: BetaManagedAgentsModelConfig`
9407
9408 Model identifier and configuration.
9409
9410
9398 - `"claude-sonnet-4-5
No line in this hunk matches that.