The whole hunk
from line 115, old and new numbered
/
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 115
115115
116116- `"anthropic-workspace-id": optional string`
117117
118 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
119
120 Only needed for credentials that can act on more than one Workspace. A credential that belongs to a specific Workspace may omit it; if sent, it must match that Workspace.
121
118122### Body parameters
119123
120124- `agent: string or BetaManagedAgentsAgentParams or BetaManagedAgentsAgentWithOverridesParams`
from line 269
265269
266270 - `"low"`
267271
272 Low effort. Favors latency over reasoning depth.
273
268274 - `"medium"`
269275
276 Medium effort. Balances latency and reasoning depth.
277
270278 - `"high"`
271279
280 High effort. Favors reasoning depth.
281
272282 - `"xhigh"`
273283
284 Extra-high effort. Not all models accept this level.
285
274286 - `"max"`
275287
288 Maximum effort. Favors reasoning depth over latency.
289
276290 - `BetaManagedAgentsEffortLow object`
277291
278292 Low effort. Favors latency over reasoning depth.
from line 1943
19291943
19301944 - `"rescheduling"`
19311945
1946 Transient error occurred, retrying automatically.
1947
19321948 - `"running"`
19331949
1950 Agent is actively executing.
1951
19341952 - `"idle"`
19351953
1954 Agent is waiting for input, including user messages or tool confirmations. Sessions start in idle.
1955
19361956 - `"terminated"`
19371957
1958 Session has ended, either due to an error or completion.
1959
19381960 - `title: string or null`
19391961
19401962 - `updated_at: string`
from line 2324
23022324
23032325 - `"rescheduling"`
23042326
2327 Transient error occurred, retrying automatically.
2328
23052329 - `"running"`
23062330
2331 Agent is actively executing.
2332
23072333 - `"idle"`
23082334
2335 Agent is waiting for input, including user messages or tool confirmations. Sessions start in idle.
2336
23092337 - `"terminated"`
23102338
2339 Session has ended, either due to an error or completion.
2340
23112341### Headers
23122342
23132343- `"anthropic-beta": optional array of AnthropicBeta`
from line 2442
24122442
24132443- `"anthropic-workspace-id": optional string`
24142444
2445 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
2446
2447 Only needed for credentials that can act on more than one Workspace. A credential that belongs to a specific Workspace may omit it; if sent, it must match that Workspace.
2448
24152449### Returns
24162450
24172451- `data: optional array of BetaManagedAgentsSession`
from line 3244
32103244
32113245 - `"rescheduling"`
32123246
3247 Transient error occurred, retrying automatically.
3248
32133249 - `"running"`
32143250
3251 Agent is actively executing.
3252
32153253 - `"idle"`
32163254
3255 Agent is waiting for input, including user messages or tool confirmations. Sessions start in idle.
3256
32173257 - `"terminated"`
32183258
3259 Session has ended, either due to an error or completion.
3260
32193261 - `title: string or null`
32203262
32213263 - `updated_at: string`
from line 3669
36273669
36283670- `"anthropic-workspace-id": optional string`
36293671
3672 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
3673
3674 Only needed for credentials that can act on more than one Workspace. A credential that belongs to a specific Workspace may omit it; if sent, it must match that Workspace.
3675
36303676### Returns
36313677
36323678- `BetaManagedAgentsSession object`
from line 4471
44254471
44264472 - `"rescheduling"`
44274473
4474 Transient error occurred, retrying automatically.
4475
44284476 - `"running"`
44294477
4478 Agent is actively executing.
4479
44304480 - `"idle"`
44314481
4482 Agent is waiting for input, including user messages or tool confirmations. Sessions start in idle.
4483
44324484 - `"terminated"`
44334485
4486 Session has ended, either due to an error or completion.
4487
44344488 - `title: string or null`
44354489
44364490 - `updated_at: string`
from line 4882
48284882
48294883- `"anthropic-workspace-id": optional string`
48304884
4885 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
4886
4887 Only needed for credentials that can act on more than one Workspace. A credential that belongs to a specific Workspace may omit it; if sent, it must match that Workspace.
4888
48314889### Body parameters
48324890
48334891- `agent: optional BetaManagedAgentsSessionAgentUpdate`
from line 6170
61126170
61136171 - `"rescheduling"`
61146172
6173 Transient error occurred, retrying automatically.
6174
61156175 - `"running"`
61166176
6177 Agent is actively executing.
6178
61176179 - `"idle"`
61186180
6181 Agent is waiting for input, including user messages or tool confirmations. Sessions start in idle.
6182
61196183 - `"terminated"`
61206184
6185 Session has ended, either due to an error or completion.
6186
61216187 - `title: string or null`
61226188
61236189 - `updated_at: string`
from line 6585
65196585
65206586- `"anthropic-workspace-id": optional string`
65216587
6588 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
6589
6590 Only needed for credentials that can act on more than one Workspace. A credential that belongs to a specific Workspace may omit it; if sent, it must match that Workspace.
6591
65226592### Returns
65236593
65246594- `BetaManagedAgentsDeletedSession object`
from line 6732
66626732
66636733- `"anthropic-workspace-id": optional string`
66646734
6735 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
6736
6737 Only needed for credentials that can act on more than one Workspace. A credential that belongs to a specific Workspace may omit it; if sent, it must match that Workspace.
6738
66656739### Returns
66666740
66676741- `BetaManagedAgentsSession object`
from line 7534
74607534
74617535 - `"rescheduling"`
74627536
7537 Transient error occurred, retrying automatically.
7538
74637539 - `"running"`
74647540
7541 Agent is actively executing.
7542
74657543 - `"idle"`
74667544
7545 Agent is waiting for input, including user messages or tool confirmations. Sessions start in idle.
7546
74677547 - `"terminated"`
74687548
7549 Session has ended, either due to an error or completion.
7550
74697551 - `title: string or null`
74707552
74717553 - `updated_at: string`
from line 8014
79328014
79338015 - `"low"`
79348016
8017 Low effort. Favors latency over reasoning depth.
8018
79358019 - `"medium"`
79368020
8021 Medium effort. Balances latency and reasoning depth.
8022
79378023 - `"high"`
79388024
8025 High effort. Favors reasoning depth.
8026
79398027 - `"xhigh"`
79408028
8029 Extra-high effort. Not all models accept this level.
8030
79418031 - `"max"`
79428032
8033 Maximum effort. Favors reasoning depth over latency.
8034
79438035 - `BetaManagedAgentsEffortLow object`
79448036
79458037 Low effort. Favors latency over reasoning depth.
from line 8436
83448436
83458437 Default configuration for all tools in a toolset.
83468438
8347 - `enabled: optional boolean or null`
8348
8349 Whether tools are enabled and available to Claude by default. Defaults to true if not specified.
8350
8351 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
8352
8353 Permission policy for tool execution.
8354
8355 - `BetaManagedAgentsAlwaysAllowPolicy object`
8356
8357 Tool calls are automatically approved without user confirmation.
8358
8359 - `BetaManagedAgentsAlwaysAskPolicy object`
8360
8361 Tool calls require user confirmation before execution.
8362
8363 - `BetaManagedAgentsAutoPolicy object`
8364
8365 The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
8366
8367 - `BetaManagedAgentsMCPToolsetParams object`
8368
8369 Configuration for tools from an MCP server defined in `mcp_servers`.
8370
8371 - `type: "mcp_toolset"`
8372
8373 - `mcp_server_name: string`
8374
8375 Name of the MCP server. Must match a server name from the mcp_servers array. 1-255 characters.
8376
8377 minLength: 1, maxLength: 255
8378
8379 - `configs: optional array of BetaManagedAgentsMCPToolConfigParams`
8380
8381 Per-tool configuration overrides.
8382
8383 - `name: string`
8384
8385 Name of the MCP tool to configure. 1-128 characters.
8386
8387 minLength: 1, maxLength: 128
8388
8389 - `enabled: optional boolean or null`
8390
8391 Whether this tool is enabled. Overrides the `default_config` setting.
8392
8393 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
8394
8395 Permission policy for tool execution.
8396
8397 - `BetaManagedAgentsAlwaysAllowPolicy object`
8398
8399 Tool calls are automatically approved without user confirmation.
8400
8401 - `BetaManagedAgentsAlwaysAskPolicy object`
8402
8403 Tool calls require user confirmation before execution.
8404
8405 - `BetaManagedAgentsAutoPolicy object`
8406
8407 The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
8408
8409 - `default_config: optional BetaManagedAgentsMCPToolsetDefaultConfigParams or null`
8410
8411 Default configuration for all tools from an MCP server.
8412
8413 - `enabled: optional boolean or null`
8414
8415 Whether tools are enabled by default. Defaults to true if not specified.
8416
8417 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
8418
8419 Permission policy for tool execution.
8420
8421 - `BetaManagedAgentsAlwaysAllowPolicy object`
8422
8423 Tool calls are automatically approved without user confirmation.
8424
8425 - `BetaManagedAgentsAlwaysAskPolicy object`
8426
8427 Tool calls require user confirmation before execution.
8428
8429 - `BetaManagedAgentsAutoPolicy object`
8430
8431 The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
8432
8433 - `BetaManagedAgentsCustomToolParams object`
8434
8435 A custom tool that is executed by the API client rather than the agent. When the agent calls this tool, an `agent.custom_tool_use` event is emitted and the session goes idle, waiting for the client to provide the result via a `user.custom_tool_result` event.
8436
8437 - `type: "custom"`
8438
8439 - `description: string`
8440
8441 Description of what the tool does, shown to the agent to hel
8439