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

Versions changed

api/beta/agents/versions

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: Versions
3url: https://platform.claude.com/docs/en/api/beta/agents/versions
4---
5 
16# Versions
27 
38## List Agent Versions
from line 35
3035 
3136 - `string`
3237 
33 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 41 more`
38 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 42 more`
3439 
3540 - `"message-batches-2024-09-24"`
3641 
from line 83
7883 
7984 - `"user-profiles-2026-08-18"`
8085 
86 - `"user-profiles-2026-09-04"`
87 
8188 - `"advisor-tool-2026-03-01"`
8289 
8390 - `"managed-agents-2026-04-01"`
from line 127
120127 
121128 - `"mid-conversation-system-clear-at-2026-08-21"`
122129 
130- `"anthropic-workspace-id": optional string`
131 
123132### Returns
124133 
125134- `data: array of BetaManagedAgentsAgent`
from line 135
126135 
127136 Agent versions.
128137 
138 - `type: "agent"`
139 
129140 - `id: string`
130141 
131142 - `archived_at: string or null`
from line 155
144155 
145156 - `mcp_servers: array of BetaManagedAgentsMCPServerURLDefinition`
146157 
147 - `name: string`
148 
149158 - `type: "url"`
150159 
160 - `name: string`
161 
151162 - `url: string`
152163 
153164 - `metadata: map[string]`
from line 287
276287 
277288 Resolved coordinator topology with a concrete agent roster.
278289 
290 - `type: "coordinator"`
291 
279292 - `agents: array of BetaManagedAgentsAgentReference or BetaManagedAgentsAdvisor`
280293 
281294 Agents the coordinator may spawn as session threads, each resolved to a specific version.
from line 297
284297 
285298 A resolved agent reference with a concrete version.
286299 
287 - `id: string`
288 
289300 - `type: "agent"`
290301 
302 - `id: string`
303 
291304 - `version: number`
292305 
293306 format: int32
from line 309
296309 
297310 Platform advisor roster entry: a model the session's primary thread may consult mid-turn.
298311 
312 - `type: "advisor"`
313 
299314 - `model: string`
300315 
301316 The advisor model id.
302317 
303 - `type: "advisor"`
304 
305 - `type: "coordinator"`
306 
307318 - `name: string`
308319 
309320 - `skills: array of BetaManagedAgentsAnthropicSkill or BetaManagedAgentsCustomSkill`
from line 323
312323 
313324 A resolved Anthropic-managed skill.
314325 
315 - `skill_id: string`
316 
317326 - `type: "anthropic"`
318327 
328 - `skill_id: string`
329 
319330 - `version: string`
320331 
321332 - `BetaManagedAgentsCustomSkill object`
from line 333
322333 
323334 A resolved user-created custom skill.
324335 
325 - `skill_id: string`
326 
327336 - `type: "custom"`
328337 
338 - `skill_id: string`
339 
329340 - `version: string`
330341 
331342 - `system: string or null`
from line 345
334345 
335346 - `BetaManagedAgentsAgentToolset20260401 object`
336347 
348 - `type: "agent_toolset_20260401"`
349 
337350 - `configs: array of BetaManagedAgentsAgentToolConfig`
338351 
339352 - `BetaManagedAgentsBashToolConfig object`
from line 353
340353 
341354 Configuration for the bash tool.
342355 
356 - `type: "bash"`
357 
343358 - `enabled: boolean`
344359 
345360 - `name: "bash"`
346361 
347 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
362 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
348363 
349364 Permission policy for tool execution.
350365 
from line 375
360375 
361376 - `type: "always_ask"`
362377 
363 - `type: "bash"`
378 - `BetaManagedAgentsAutoPolicy object`
364379 
380 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.
381 
382 - `type: "auto"`
383 
365384 - `BetaManagedAgentsEditToolConfig object`
366385 
367386 Configuration for the edit tool.
368387 
388 - `type: "edit"`
389 
369390 - `enabled: boolean`
370391 
371392 - `name: "edit"`
372393 
373 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
394 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
374395 
375396 Permission policy for tool execution.
376397 
from line 403
382403 
383404 Tool calls require user confirmation before execution.
384405 
385 - `type: "edit"`
406 - `BetaManagedAgentsAutoPolicy object`
386407 
408 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.
409 
387410 - `BetaManagedAgentsReadToolConfig object`
388411 
389412 Configuration for the read tool.
390413 
414 - `type: "read"`
415 
391416 - `enabled: boolean`
392417 
393418 - `name: "read"`
394419 
395 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
420 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
396421 
397422 Permission policy for tool execution.
398423 
from line 429
404429 
405430 Tool calls require user confirmation before execution.
406431 
407 - `type: "read"`
432 - `BetaManagedAgentsAutoPolicy object`
408433 
434 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.
435 
409436 - `BetaManagedAgentsWriteToolConfig object`
410437 
411438 Configuration for the write tool.
412439 
440 - `type: "write"`
441 
413442 - `enabled: boolean`
414443 
415444 - `name: "write"`
416445 
417 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
446 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
418447 
419448 Permission policy for tool execution.
420449 
from line 455
426455 
427456 Tool calls require user confirmation before execution.
428457 
429 - `type: "write"`
458 - `BetaManagedAgentsAutoPolicy object`
430459 
460 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.
461 
431462 - `BetaManagedAgentsGlobToolConfig object`
432463 
433464 Configuration for the glob tool.
434465 
466 - `type: "glob"`
467 
435468 - `enabled: boolean`
436469 
437470 - `name: "glob"`
438471 
439 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
472 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
440473 
441474 Permission policy for tool execution.
442475 
from line 481
448481 
449482 Tool calls require user confirmation before execution.
450483 
451 - `type: "glob"`
484 - `BetaManagedAgentsAutoPolicy object`
452485 
486 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.
487 
453488 - `BetaManagedAgentsGrepToolConfig object`
454489 
455490 Configuration for the grep tool.
456491 
492 - `type: "grep"`
493 
457494 - `enabled: boolean`
458495 
459496 - `name: "grep"`
460497 
461 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
498 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
462499 
463500 Permission policy for tool execution.
464501 
from line 507
470507 
471508 Tool calls require user confirmation before execution.
472509 
473 - `type: "grep"`
510 - `BetaManagedAgentsAutoPolicy object`
474511 
512 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.
513 
475514 - `BetaManagedAgentsWebFetchToolConfig object`
476515 
477516 Configuration for the web_fetch tool.
478517 
518 - `type: "web_fetch"`
519 
479520 - `enabled: boolean`
480521 
481522 - `name: "web_fetch"`
482523 
483 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
524 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
484525 
485526 Permission policy for tool execution.
486527 
from line 533
492533 
493534 Tool calls require user confirmation before execution.
494535 
495 - `type: "web_fetch"`
536 - `BetaManagedAgentsAutoPolicy object`
496537 
538 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.
539 
497540 - `allowed_domains: optional array of string`
498541 
499542 - `blocked_domains: optional array of string`
from line 549
506549 
507550 Configuration for the web_search tool.
508551 
552 - `type: "web_search"`
553 
509554 - `enabled: boolean`
510555 
511556 - `name: "web_search"`
512557 
513 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
558 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
514559 
515560 Permission policy for tool execution.
516561 
from line 567
522567 
523568 Tool calls require user confirmation before execution.
524569 
525 - `type: "web_search"`
570 - `BetaManagedAgentsAutoPolicy object`
526571 
572 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.
573 
527574 - `allowed_domains: optional array of string`
528575 
529576 - `blocked_domains: optional array of string`
from line 611
564611 
565612 - `enabled: boolean`
566613 
567 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
614 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
568615 
569616 Permission policy for tool execution.
570617 
from line 623
576623 
577624 Tool calls require user confirmation before execution.
578625 
579 - `type: "agent_toolset_20260401"`
626 - `BetaManagedAgentsAutoPolicy object`
580627 
628 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.
629 
581630 - `BetaManagedAgentsMCPToolset object`
582631 
632 - `type: "mcp_toolset"`
633 
583634 - `configs: array of BetaManagedAgentsMCPToolConfig`
584635 
585636 - `enabled: boolean`
from line 637
586637 
587638 - `name: string`
588639 
589 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
640 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
590641 
591642 Permission policy for tool execution.
592643 
from line 649
598649 
599650 Tool calls require user confirmation before execution.
600651 
652 - `BetaManagedAgentsAutoPolicy object`
653 
654 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.
655 
601656 - `default_config: BetaManagedAgentsMCPToolsetDefaultConfig`
602657 
603658 Resolved default configuration for all tools from an MCP server.
from line 659
604659 
605660 - `enabled: boolean`
606661 
607 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy`
662 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`
608663 
609664 Permission policy for tool execution.
610665 
from line 671
616671 
617672 Tool calls require user confirmation before execution.
618673 
619 - `mcp_server_name: string`
674 - `BetaManagedAgentsAutoPolicy object`
620675 
621 - `type: "mcp_toolset"`
676 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.
622677 
678 - `mcp_server_name: string`
679 
623680 - `BetaManagedAgentsCustomTool object`
624681 
625682 A custom tool as returned in API responses.
626683 
684 - `type: "custom"`
685 
627686 - `description: string`
628687 
629688 - `input_schema: BetaManagedAgentsCustomToolInputSchema`
from line 696
637696 - `required: optional array of string or null`
638697 
639698 - `name: string`
640 
641 - `type: "custom"`
642 
643 - `type: "agent"`
644699 
645700 - `updated_at: string`
646701