Sweep 22 Sep 2026 · 17:19Z Build v2.1.280 501 read Stable v2.1.267 Latest v2.1.280 Next v2.1.280 Feeds RSS JSON llms.txt Unofficial
One change · api

Get Agent changed

api/beta/agents/retrieve

Nearest release: v2.1.268, published under 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+368added
Lines−313removed
From line 1 where the diff opens
First seen 14 Aug 2026 this site's first read of the page
Recorded edits9to this page, all time

The whole hunk

from line 1, old and new numbered
/
lines
from line 1
1---
2title: Get Agent
3url: https://platform.claude.com/docs/en/api/beta/agents/retrieve
4---
5 
16# Get Agent
27 
38**GET** `/v1/agents/{agent_id}`
from line 29
2429 
2530 - `string`
2631 
27 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 41 more`
32 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 42 more`
2833 
2934 - `"message-batches-2024-09-24"`
3035 
from line 77
7277 
7378 - `"user-profiles-2026-08-18"`
7479 
80 - `"user-profiles-2026-09-04"`
81 
7582 - `"advisor-tool-2026-03-01"`
7683 
7784 - `"managed-agents-2026-04-01"`
from line 121
114121 
115122 - `"mid-conversation-system-clear-at-2026-08-21"`
116123 
124- `"anthropic-workspace-id": optional string`
125 
117126## Returns
118127 
119128- `BetaManagedAgentsAgent object`
from line 129
120129 
121130 A Managed Agents `agent`.
122131 
132 - `type: "agent"`
133 
123134 - `id: string`
124135 
125136 - `archived_at: string or null`
from line 149
138149 
139150 - `mcp_servers: array of BetaManagedAgentsMCPServerURLDefinition`
140151 
141 - `name: string`
142 
143152 - `type: "url"`
144153 
154 - `name: string`
155 
145156 - `url: string`
146157 
147158 - `metadata: map[string]`
from line 281
270281 
271282 Resolved coordinator topology with a concrete agent roster.
272283 
284 - `type: "coordinator"`
285 
273286 - `agents: array of BetaManagedAgentsAgentReference or BetaManagedAgentsAdvisor`
274287 
275288 Agents the coordinator may spawn as session threads, each resolved to a specific version.
from line 291
278291 
279292 A resolved agent reference with a concrete version.
280293 
281 - `id: string`
282 
283294 - `type: "agent"`
284295 
296 - `id: string`
297 
285298 - `version: number`
286299 
287300 format: int32
from line 303
290303 
291304 Platform advisor roster entry: a model the session's primary thread may consult mid-turn.
292305 
306 - `type: "advisor"`
307 
293308 - `model: string`
294309 
295310 The advisor model id.
296311 
297 - `type: "advisor"`
298 
299 - `type: "coordinator"`
300 
301312 - `name: string`
302313 
303314 - `skills: array of BetaManagedAgentsAnthropicSkill or BetaManagedAgentsCustomSkill`
from line 317
306317 
307318 A resolved Anthropic-managed skill.
308319 
309 - `skill_id: string`
310 
311320 - `type: "anthropic"`
312321 
322 - `skill_id: string`
323 
313324 - `version: string`
314325 
315326 - `BetaManagedAgentsCustomSkill object`
from line 327
316327 
317328 A resolved user-created custom skill.
318329 
319 - `skill_id: string`
320 
321330 - `type: "custom"`
322331 
332 - `skill_id: string`
333 
323334 - `version: string`
324335 
325336 - `system: string or null`
from line 339
328339 
329340 - `BetaManagedAgentsAgentToolset20260401 object`
330341 
342 - `type: "agent_toolset_20260401"`
343 
331344 - `configs: array of BetaManagedAgentsAgentToolConfig`
332345 
333346 - `BetaManagedAgentsBashToolConfig object`
from line 347
334347 
335348 Configuration for the bash tool.
336349 
350 - `type: "bash"`
351 
337352 - `enabled: boolean`
338353 
339354 - `name: "bash"`
340355 
341 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
356 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
342357 
343358 Permission policy for tool execution.
344359 
from line 369
354369 
355370 - `type: "always_ask"`
356371 
357 - `type: "bash"`
372 - `BetaManagedAgentsAutoPolicy object`
358373 
374 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.
375 
376 - `type: "auto"`
377 
359378 - `BetaManagedAgentsEditToolConfig object`
360379 
361380 Configuration for the edit tool.
362381 
382 - `type: "edit"`
383 
363384 - `enabled: boolean`
364385 
365386 - `name: "edit"`
366387 
367 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
388 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
368389 
369390 Permission policy for tool execution.
370391 
from line 397
376397 
377398 Tool calls require user confirmation before execution.
378399 
379 - `type: "edit"`
400 - `BetaManagedAgentsAutoPolicy object`
380401 
402 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.
403 
381404 - `BetaManagedAgentsReadToolConfig object`
382405 
383406 Configuration for the read tool.
384407 
408 - `type: "read"`
409 
385410 - `enabled: boolean`
386411 
387412 - `name: "read"`
388413 
389 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
414 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
390415 
391416 Permission policy for tool execution.
392417 
from line 423
398423 
399424 Tool calls require user confirmation before execution.
400425 
401 - `type: "read"`
426 - `BetaManagedAgentsAutoPolicy object`
402427 
428 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.
429 
403430 - `BetaManagedAgentsWriteToolConfig object`
404431 
405432 Configuration for the write tool.
406433 
434 - `type: "write"`
435 
407436 - `enabled: boolean`
408437 
409438 - `name: "write"`
410439 
411 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
440 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
412441 
413442 Permission policy for tool execution.
414443 
from line 449
420449 
421450 Tool calls require user confirmation before execution.
422451 
423 - `type: "write"`
452 - `BetaManagedAgentsAutoPolicy object`
424453 
454 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.
455 
425456 - `BetaManagedAgentsGlobToolConfig object`
426457 
427458 Configuration for the glob tool.
428459 
460 - `type: "glob"`
461 
429462 - `enabled: boolean`
430463 
431464 - `name: "glob"`
432465 
433 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
466 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
434467 
435468 Permission policy for tool execution.
436469 
from line 475
442475 
443476 Tool calls require user confirmation before execution.
444477 
445 - `type: "glob"`
478 - `BetaManagedAgentsAutoPolicy object`
446479 
480 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.
481 
447482 - `BetaManagedAgentsGrepToolConfig object`
448483 
449484 Configuration for the grep tool.
450485 
486 - `type: "grep"`
487 
451488 - `enabled: boolean`
452489 
453490 - `name: "grep"`
454491 
455 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
492 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
456493 
457494 Permission policy for tool execution.
458495 
from line 501
464501 
465502 Tool calls require user confirmation before execution.
466503 
467 - `type: "grep"`
504 - `BetaManagedAgentsAutoPolicy object`
468505 
506 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.
507 
469508 - `BetaManagedAgentsWebFetchToolConfig object`
470509 
471510 Configuration for the web_fetch tool.
472511 
512 - `type: "web_fetch"`
513 
473514 - `enabled: boolean`
474515 
475516 - `name: "web_fetch"`
476517 
477 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
518 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
478519 
479520 Permission policy for tool execution.
480521 
from line 527
486527 
487528 Tool calls require user confirmation before execution.
488529 
489 - `type: "web_fetch"`
530 - `BetaManagedAgentsAutoPolicy object`
490531 
532 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.
533 
491534 - `allowed_domains: optional array of string`
492535 
493536 - `blocked_domains: optional array of string`
from line 543
500543 
501544 Configuration for the web_search tool.
502545 
546 - `type: "web_search"`
547 
503548 - `enabled: boolean`
504549 
505550 - `name: "web_search"`
506551 
507 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
552 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
508553 
509554 Permission policy for tool execution.
510555 
from line 561
516561 
517562 Tool calls require user confirmation before execution.
518563 
519 - `type: "web_search"`
564 - `BetaManagedAgentsAutoPolicy object`
520565 
566 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.
567 
521568 - `allowed_domains: optional array of string`
522569 
523570 - `blocked_domains: optional array of string`
from line 605
558605 
559606 - `enabled: boolean`
560607 
561 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
608 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
562609 
563610 Permission policy for tool execution.
564611 
from line 617
570617 
571618 Tool calls require user confirmation before execution.
572619 
573 - `type: "agent_toolset_20260401"`
620 - `BetaManagedAgentsAutoPolicy object`
574621 
622 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.
623 
575624 - `BetaManagedAgentsMCPToolset object`
576625 
626 - `type: "mcp_toolset"`
627 
577628 - `configs: array of BetaManagedAgentsMCPToolConfig`
578629 
579630 - `enabled: boolean`
from line 631
580631 
581632 - `name: string`
582633 
583 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
634 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
584635 
585636 Permission policy for tool execution.
586637 
from line 643
592643 
593644 Tool calls require user confirmation before execution.
594645 
646 - `BetaManagedAgentsAutoPolicy object`
647 
648 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.
649 
595650 - `default_config: BetaManagedAgentsMCPToolsetDefaultConfig`
596651 
597652 Resolved default configuration for all tools from an MCP server.
from line 653
598653 
599654 - `enabled: boolean`
600655 
601 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
656 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
602657 
603658 Permission policy for tool execution.
604659 
from line 665
610665 
611666 Tool calls require user confirmation before execution.
612667 
613 - `mcp_server_name: string`
668 - `BetaManagedAgentsAutoPolicy object`
614669 
615 - `type: "mcp_toolset"`
670 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.
616671 
672 - `mcp_server_name: string`
673 
617674 - `BetaManagedAgentsCustomTool object`
618675 
619676 A custom tool as returned in API responses.
620677 
678 - `type: "custom"`
679 
621680 - `description: string`
622681 
623682 - `input_schema: BetaManagedAgentsCustomToolInputSchema`
from line 690
631690 - `required: optional array of string or null`
632691 
633692 - `name: string`
634 
635 - `type: "custom"`
636 
637 - `type: "agent"`
638693 
639694 - `updated_at: string`
640695