Follow Discord
Sweep 25 Sep 2026 · 19:33Z Build v2.1.283 504 read Stable v2.1.274 Latest v2.1.283 Next v2.1.283 Feeds RSS JSON llms.txt llms-full.txt Unofficial
One change · api

Create Session changedapi/beta/sessions/create

Nearest release: v2.1.283, published 3 hours 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+1,738added
Lines−1,738removed
From line 181 where the diff opens
First seen 14 Aug 2026 this site's first read of the page
Recorded edits11to this page, all time

The whole hunk

from line 181, old and new numbered
/
lines
from line 181
181181 
182182 Replacement model. Accepts the model string, e.g. `claude-opus-5`, or a `model_config` object. Omit to use the agent's model.
183183 
184 - `BetaManagedAgentsModel = "claude-opus-5-5" or "claude-fable-5-1" or "claude-sonnet-5" or 12 more or string`
184 - `BetaManagedAgentsModel = string or "claude-opus-5-5" or "claude-fable-5-1" or "claude-sonnet-5" or 12 more`
185185 
186186 The model that will power your agent.
187187 
188188 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
189189 
190 - `string`
191 
190192 - `"claude-opus-5-5" or "claude-fable-5-1" or "claude-sonnet-5" or 12 more`
191193 
192194 The model that will power your agent.
from line 255
253255 
254256 High-performance model for agents and coding
255257 
256 - `string`
257 
258258 - `BetaManagedAgentsModelConfigParams object`
259259 
260260 An object that defines additional configuration control over model use
from line 329
329329 
330330 - `speed: optional "standard" or "fast" or null`
331331 
332 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.
332 Inference speed mode. Defaults to `standard`.
333333 
334334 - `"standard"`
335335 
from line 411
411411 
412412 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
413413 
414 Permission policy for tool execution.
414 Permission policy for this tool. Controls whether tool calls are auto-approved or require confirmation.
415415 
416416 - `BetaManagedAgentsAlwaysAllowPolicy object`
417417 
from line 447
447447 
448448 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
449449 
450 Permission policy for tool execution.
450 Permission policy for this tool. Controls whether tool calls are auto-approved or require confirmation.
451451 
452452 - `BetaManagedAgentsAlwaysAllowPolicy object`
453453 
from line 477
477477 
478478 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
479479 
480 Permission policy for tool execution.
480 Permission policy for this tool. Controls whether tool calls are auto-approved or require confirmation.
481481 
482482 - `BetaManagedAgentsAlwaysAllowPolicy object`
483483 
from line 507
507507 
508508 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
509509 
510 Permission policy for tool execution.
510 Permission policy for this tool. Controls whether tool calls are auto-approved or require confirmation.
511511 
512512 - `BetaManagedAgentsAlwaysAllowPolicy object`
513513 
from line 537
537537 
538538 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
539539 
540 Permission policy for tool execution.
540 Permission policy for this tool. Controls whether tool calls are auto-approved or require confirmation.
541541 
542542 - `BetaManagedAgentsAlwaysAllowPolicy object`
543543 
from line 567
567567 
568568 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
569569 
570 Permission policy for tool execution.
570 Permission policy for this tool. Controls whether tool calls are auto-approved or require confirmation.
571571 
572572 - `BetaManagedAgentsAlwaysAllowPolicy object`
573573 
from line 611
611611 
612612 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
613613 
614 Permission policy for tool execution.
614 Permission policy for this tool. Controls whether tool calls are auto-approved or require confirmation.
615615 
616616 - `BetaManagedAgentsAlwaysAllowPolicy object`
617617 
from line 649
649649 
650650 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
651651 
652 Permission policy for tool execution.
652 Permission policy for this tool. Controls whether tool calls are auto-approved or require confirmation.
653653 
654654 - `BetaManagedAgentsAlwaysAllowPolicy object`
655655 
from line 695
695695 
696696 - `default_config: optional BetaManagedAgentsAgentToolsetDefaultConfigParams or null`
697697 
698 Default configuration for all tools in a toolset.
698 Default configuration applied to all tools in this set.
699699 
700700 - `enabled: optional boolean or null`
701701 
from line 703
703703 
704704 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
705705 
706 Permission policy for tool execution.
706 Default permission policy for tools. Controls whether tool calls are auto-approved or require confirmation.
707707 
708708 - `BetaManagedAgentsAlwaysAllowPolicy object`
709709 
from line 745
745745 
746746 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
747747 
748 Permission policy for tool execution.
748 Permission policy for this tool. Overrides the `default_config` setting.
749749 
750750 - `BetaManagedAgentsAlwaysAllowPolicy object`
751751 
from line 761
761761 
762762 - `default_config: optional BetaManagedAgentsMCPToolsetDefaultConfigParams or null`
763763 
764 Default configuration for all tools from an MCP server.
764 Default configuration for all tools from this server.
765765 
766766 - `enabled: optional boolean or null`
767767 
from line 769
769769 
770770 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`
771771 
772 Permission policy for tool execution.
772 Default permission policy for tools from this server.
773773 
774774 - `BetaManagedAgentsAlwaysAllowPolicy object`
775775 
from line 797
797797 
798798 - `input_schema: BetaManagedAgentsCustomToolInputSchema`
799799 
800 JSON Schema for custom tool input parameters.
800 JSON Schema defining the expected input parameters for the tool.
801801 
802802 - `type: "object"`
803803 
from line 825
825825 
826826- `budget: optional BetaManagedAgentsBudgetLimit`
827827 
828 A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches `max_list_cost`.
828 Enforced spend ceiling for the session. Omit to create an uncapped session. Every model the session can run — the agent's model and each callable agent's model — must have a public list price, or the request is rejected with reason `model_not_budgetable`.
829829 
830830 - `type: "limit"`
831831 
832832 - `max_list_cost: BetaMonetaryAmount`
833833 
834 A monetary amount in a specific currency.
834 Maximum list cost the session may accrue. List price is used regardless of any negotiated discount, so the cap fires at or before the actual charge.
835835 
836836 - `amount: string`
837837 
from line 875
875875 
876876 - `source: BetaManagedAgentsBase64ImageSource or BetaManagedAgentsURLImageSource or BetaManagedAgentsFileImageSource`
877877 
878 Union type for image source variants.
878 The source of the image data.
879879 
880880 - `BetaManagedAgentsBase64ImageSource object`
881881 
from line 927
927927 
928928 - `source: BetaManagedAgentsBase64DocumentSource or BetaManagedAgentsPlainTextDocumentSource or BetaManagedAgentsURLDocumentSource or BetaManagedAgentsFileDocumentSource`
929929 
930 Union type for document source variants.
930 The source of the document data.
931931 
932932 - `BetaManagedAgentsBase64DocumentSource object`
933933 
from line 1013
10131013 
10141014 - `rubric: BetaManagedAgentsFileRubricParams or BetaManagedAgentsTextRubricParams`
10151015 
1016 Rubric for grading the quality of an outcome.
1016 How to grade the outcome. Text or file reference.
10171017 
10181018 - `BetaManagedAgentsFileRubricParams object`
10191019 
from line 1129
11291129 
11301130 - `access: optional "read_write" or "read_only" or null`
11311131 
1132 Access mode for an attached memory store.
1132 Access mode for the mounted store. Defaults to read_write. read_only mounts the store as a read-only filesystem.
11331133 
11341134 - `"read_write"`
11351135 
from line 1189
11891189 
11901190 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
11911191 
1192 - `string`
1193 
11921194 - `"claude-opus-5-5" or "claude-fable-5-1" or "claude-sonnet-5" or 12 more`
11931195 
11941196 The model that will power your agent.
from line 1257
12551257 
12561258 High-performance model for agents and coding
12571259 
1258 - `string`
1259 
12601260 - `effort: optional BetaManagedAgentsEffortLow or BetaManagedAgentsEffortMedium or BetaManagedAgentsEffortHigh or 2 more`
12611261 
1262 How hard Claude works on each turn. Sets `output_config.effort` on every Messages call the session makes.
1262 How hard Claude works on each inference call. One of `low`, `medium`, `high`, `xhigh`, `max`. Always present; resolved to the per-model default at save time when not supplied.
12631263 
12641264 - `BetaManagedAgentsEffortLow object`
12651265 
from line 1297
12971297 
12981298 - `speed: optional "standard" or "fast"`
12991299 
1300 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.
1300 Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Defaults to `standard`. Not all models support `fast`; invalid combinations are rejected at create time.
13011301 
13021302 - `"standard"`
13031303 
from line 1305
13051305 
13061306 - `multiagent: BetaManagedAgentsSessionMultiagentCoordinator or null`
13071307 
1308 Resolved coordinator topology with full agent definitions for each roster member.
1308 Resolved multiagent orchestration configuration. Null when the agent is single-threaded.
13091309 
13101310 - `type: "coordinator"`
13111311 
from line 1761
17611761 
17621762 - `archived_at: string or null`
17631763 
1764 A timestamp in RFC 3339 format
1764 When the session was archived. Null if not archived.
17651765 
17661766 format: date-time
17671767 
17681768 - `budget: BetaManagedAgentsBudgetLimit or null`
17691769 
1770 A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches `max_list_cost`.
1770 The session's enforced spend ceiling, or null when no budget is set.
17711771 
17721772 - `type: "limit"`
17731773 
17741774 - `max_list_cost: BetaMonetaryAmount`
17751775 
1776 A monetary amount in a specific currency.
1776 Maximum list cost the session may accrue. List price is used regardless of any negotiated discount, so the cap fires at or before the actual charge.
17771777 
17781778 - `amount: string`
17791779 
from line 1801
18011801 
18021802 - `completed_at: string or null`
18031803 
1804 A timestamp in RFC 3339 format
1804 When the outcome reached a terminal result. Null while `pending`/`running`/`evaluating`.
18051805 
18061806 format: date-time
18071807 
from line 1907
19071907 
19081908 - `access: optional "read_write" or "read_only" or null`
19091909 
1910 Access mode for an attached memory store.
1910 Access mode for the mounted store. Defaults to `read_write`. `read_only` mounts the store as a read-only filesystem.
19111911 
19121912 - `"read_write"`
19131913 
from line 1933
19331933 
19341934 - `stats: BetaManagedAgentsSessionStats`
19351935 
1936 Timing statistics for a session.
1936 Timing statistics for the session.
19371937 
19381938 - `active_seconds: optional number`
19391939 
from line 1977
19771977 
19781978 - `usage: BetaManagedAgentsSessionUsage`
19791979 
1980 Cumulative token usage for a session across all turns.
1980 Cumulative token usage for the session.
19811981 
19821982 - `active_seconds: optional number`
19831983 
from line 1987
19871987 
19881988 - `cache_creation: optional BetaManagedAgentsCacheCreationUsage`
19891989 
1990 Prompt-cache creation token usage broken down by cache lifetime.
1990 Tokens used to create prompt cache entries, broken down by cache TTL.
19911991 
19921992 - `ephemeral_1h_input_tokens: optional number`
19931993 
from line 2015
20152015 
20162016 - `list_cost: optional BetaMonetaryAmount or null`
20172017 
2018 A monetary amount in a specific currency.
2018 Cumulative list cost of the session across all turns, priced at public list rates. Absent until cost tracking is available for the session.
20192019 
20202020 - `output_tokens: optional number`
20212021 
from line 2025
20252025 
20262026 - `server_tool_use: optional BetaManagedAgentsServerToolUsage or null`
20272027 
2028 Cumulative count of server-executed tool invocations, broken down by tool.
2028 Cumulative server-executed tool usage across all turns. Absent until server-tool tracking is available for the session.
20292029 
20302030 - `web_fetch_requests: optional number`
20312031 
Feedback