Repository Public
1---2title: Update Session3url: https://platform.claude.com/docs/en/api/beta/sessions/update4---56# Update Session78**POST** `/v1/sessions/{session_id}`910Update Session1112## Path parameters1314- `session_id: string`1516## Headers1718- `"anthropic-beta": optional array of AnthropicBeta`1920 Optional header to specify the beta version(s) you want to use.2122 - `string`2324 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 43 more`2526 - `"message-batches-2024-09-24"`2728 - `"prompt-caching-2024-07-31"`2930 - `"computer-use-2024-10-22"`3132 - `"computer-use-2025-01-24"`3334 - `"pdfs-2024-09-25"`3536 - `"token-counting-2024-11-01"`3738 - `"token-efficient-tools-2025-02-19"`3940 - `"output-128k-2025-02-19"`4142 - `"files-api-2025-04-14"`4344 - `"mcp-client-2025-04-04"`4546 - `"mcp-client-2025-11-20"`4748 - `"dev-full-thinking-2025-05-14"`4950 - `"interleaved-thinking-2025-05-14"`5152 - `"code-execution-2025-05-22"`5354 - `"extended-cache-ttl-2025-04-11"`5556 - `"context-1m-2025-08-07"`5758 - `"context-management-2025-06-27"`5960 - `"model-context-window-exceeded-2025-08-26"`6162 - `"skills-2025-10-02"`6364 - `"fast-mode-2026-02-01"`6566 - `"output-300k-2026-03-24"`6768 - `"user-profiles-2026-03-24"`6970 - `"user-profiles-2026-08-18"`7172 - `"user-profiles-2026-09-04"`7374 - `"advisor-tool-2026-03-01"`7576 - `"managed-agents-2026-04-01"`7778 - `"cache-diagnosis-2026-04-07"`7980 - `"dreaming-2026-04-21"`8182 - `"thinking-token-count-2026-05-13"`8384 - `"server-side-fallback-2026-06-01"`8586 - `"server-side-fallback-2026-07-01"`8788 - `"fallback-credit-2026-06-01"`8990 - `"fallback-credit-2026-07-01"`9192 - `"agent-memory-2026-07-22"`9394 - `"mid-conversation-tool-changes-2026-07-01"`9596 - `"compact-2026-01-12"`9798 - `"computer-use-2025-11-24"`99100 - `"mcp-tunnels-2026-06-22"`101102 - `"structured-outputs-2025-11-13"`103104 - `"task-budgets-2026-03-13"`105106 - `"thinking-display-updates-2026-08-18"`107108 - `"ce-user-management-2026-07-13"`109110 - `"mid-conversation-output-config-2026-07-01"`111112 - `"thinking-binding-controls-2026-08-01"`113114 - `"mid-conversation-system-clear-at-2026-08-21"`115116 - `"compact-2026-09-04"`117118- `"anthropic-workspace-id": optional string`119120## Body parameters121122- `agent: optional BetaManagedAgentsSessionAgentUpdate`123124 Mid-session agent configuration update. Only `tools` and `mcp_servers` are updatable. Full replacement: the provided array becomes the new value. To preserve existing entries, GET the session, modify the array, and POST it back.125126 - `mcp_servers: optional array of BetaManagedAgentsURLMCPServerParams`127128 Replacement MCP server list. Full replacement: the provided array becomes the new value. Send an empty array to clear; omit to preserve.129130 - `type: "url"`131132 - `name: string`133134 Unique name for this server, referenced by mcp_toolset configurations. 1-255 characters.135136 minLength: 1, maxLength: 255137138 - `url: string`139140 Endpoint URL for the MCP server.141142 maxLength: 2048143144 - `tools: optional array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams`145146 Replacement tool list. Full replacement: the provided array becomes the new value. Send an empty array to clear; omit to preserve.147148 - `BetaManagedAgentsAgentToolset20260401Params object`149150 Configuration for built-in agent tools. Use this to enable or disable groups of tools available to the agent.151152 - `type: "agent_toolset_20260401"`153154 - `configs: optional array of BetaManagedAgentsAgentToolConfigParams`155156 Per-tool configuration overrides.157158 - `BetaManagedAgentsBashToolConfigParams object`159160 Configuration override for the bash tool.161162 - `type: optional "bash"`163164 - `name: "bash"`165166 Must be "bash".167168 - `enabled: optional boolean or null`169170 Whether this tool is enabled and available to Claude. Overrides the default_config setting.171172 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`173174 Permission policy for tool execution.175176 - `BetaManagedAgentsAlwaysAllowPolicy object`177178 Tool calls are automatically approved without user confirmation.179180 - `type: "always_allow"`181182 - `BetaManagedAgentsAlwaysAskPolicy object`183184 Tool calls require user confirmation before execution.185186 - `type: "always_ask"`187188 - `BetaManagedAgentsAutoPolicy object`189190 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.191192 - `type: "auto"`193194 - `BetaManagedAgentsEditToolConfigParams object`195196 Configuration override for the edit tool.197198 - `type: optional "edit"`199200 - `name: "edit"`201202 Must be "edit".203204 - `enabled: optional boolean or null`205206 Whether this tool is enabled and available to Claude. Overrides the default_config setting.207208 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`209210 Permission policy for tool execution.211212 - `BetaManagedAgentsAlwaysAllowPolicy object`213214 Tool calls are automatically approved without user confirmation.215216 - `BetaManagedAgentsAlwaysAskPolicy object`217218 Tool calls require user confirmation before execution.219220 - `BetaManagedAgentsAutoPolicy object`221222 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.223224 - `BetaManagedAgentsReadToolConfigParams object`225226 Configuration override for the read tool.227228 - `type: optional "read"`229230 - `name: "read"`231232 Must be "read".233234 - `enabled: optional boolean or null`235236 Whether this tool is enabled and available to Claude. Overrides the default_config setting.237238 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`239240 Permission policy for tool execution.241242 - `BetaManagedAgentsAlwaysAllowPolicy object`243244 Tool calls are automatically approved without user confirmation.245246 - `BetaManagedAgentsAlwaysAskPolicy object`247248 Tool calls require user confirmation before execution.249250 - `BetaManagedAgentsAutoPolicy object`251252 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.253254 - `BetaManagedAgentsWriteToolConfigParams object`255256 Configuration override for the write tool.257258 - `type: optional "write"`259260 - `name: "write"`261262 Must be "write".263264 - `enabled: optional boolean or null`265266 Whether this tool is enabled and available to Claude. Overrides the default_config setting.267268 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`269270 Permission policy for tool execution.271272 - `BetaManagedAgentsAlwaysAllowPolicy object`273274 Tool calls are automatically approved without user confirmation.275276 - `BetaManagedAgentsAlwaysAskPolicy object`277278 Tool calls require user confirmation before execution.279280 - `BetaManagedAgentsAutoPolicy object`281282 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.283284 - `BetaManagedAgentsGlobToolConfigParams object`285286 Configuration override for the glob tool.287288 - `type: optional "glob"`289290 - `name: "glob"`291292 Must be "glob".293294 - `enabled: optional boolean or null`295296 Whether this tool is enabled and available to Claude. Overrides the default_config setting.297298 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`299300 Permission policy for tool execution.301302 - `BetaManagedAgentsAlwaysAllowPolicy object`303304 Tool calls are automatically approved without user confirmation.305306 - `BetaManagedAgentsAlwaysAskPolicy object`307308 Tool calls require user confirmation before execution.309310 - `BetaManagedAgentsAutoPolicy object`311312 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.313314 - `BetaManagedAgentsGrepToolConfigParams object`315316 Configuration override for the grep tool.317318 - `type: optional "grep"`319320 - `name: "grep"`321322 Must be "grep".323324 - `enabled: optional boolean or null`325326 Whether this tool is enabled and available to Claude. Overrides the default_config setting.327328 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`329330 Permission policy for tool execution.331332 - `BetaManagedAgentsAlwaysAllowPolicy object`333334 Tool calls are automatically approved without user confirmation.335336 - `BetaManagedAgentsAlwaysAskPolicy object`337338 Tool calls require user confirmation before execution.339340 - `BetaManagedAgentsAutoPolicy object`341342 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.343344 - `BetaManagedAgentsWebFetchToolConfigParams object`345346 Configuration override for the web_fetch tool.347348 - `type: optional "web_fetch"`349350 - `name: "web_fetch"`351352 Must be "web_fetch".353354 - `allowed_domains: optional array of string`355356 Only fetch URLs whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like "docs.example.com" (no scheme, port, or path). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with blocked_domains.357358 - `blocked_domains: optional array of string`359360 Never fetch URLs whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like "ads.example.com" (no scheme, port, or path). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with allowed_domains.361362 - `enabled: optional boolean or null`363364 Whether this tool is enabled and available to Claude. Overrides the default_config setting.365366 - `max_content_tokens: optional number or null`367368 Maximum number of tokens of fetched text content to include in context per call. Does not apply to binary content such as PDFs.369370 format: int32371372 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`373374 Permission policy for tool execution.375376 - `BetaManagedAgentsAlwaysAllowPolicy object`377378 Tool calls are automatically approved without user confirmation.379380 - `BetaManagedAgentsAlwaysAskPolicy object`381382 Tool calls require user confirmation before execution.383384 - `BetaManagedAgentsAutoPolicy object`385386 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.387388 - `BetaManagedAgentsWebSearchToolConfigParams object`389390 Configuration override for the web_search tool.391392 - `type: optional "web_search"`393394 - `name: "web_search"`395396 Must be "web_search".397398 - `allowed_domains: optional array of string`399400 Only return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like "docs.example.com" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with blocked_domains.401402 - `blocked_domains: optional array of string`403404 Never return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like "ads.example.com" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with allowed_domains.405406 - `enabled: optional boolean or null`407408 Whether this tool is enabled and available to Claude. Overrides the default_config setting.409410 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`411412 Permission policy for tool execution.413414 - `BetaManagedAgentsAlwaysAllowPolicy object`415416 Tool calls are automatically approved without user confirmation.417418 - `BetaManagedAgentsAlwaysAskPolicy object`419420 Tool calls require user confirmation before execution.421422 - `BetaManagedAgentsAutoPolicy object`423424 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.425426 - `user_location: optional BetaManagedAgentsUserLocation or null`427428 Approximate user location for search result localization.429430 - `type: "approximate"`431432 Location precision. Only "approximate" is supported.433434 - `city: optional string or null`435436 City name.437438 minLength: 1, maxLength: 255439440 - `country: optional string or null`441442 Two-letter ISO 3166-1 country code, uppercase.443444 - `region: optional string or null`445446 Region or state name.447448 minLength: 1, maxLength: 255449450 - `timezone: optional string or null`451452 IANA timezone identifier, e.g. "America/Los_Angeles".453454 minLength: 1, maxLength: 255455456 - `default_config: optional BetaManagedAgentsAgentToolsetDefaultConfigParams or null`457458 Default configuration for all tools in a toolset.459460 - `enabled: optional boolean or null`461462 Whether tools are enabled and available to Claude by default. Defaults to true if not specified.463464 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`465466 Permission policy for tool execution.467468 - `BetaManagedAgentsAlwaysAllowPolicy object`469470 Tool calls are automatically approved without user confirmation.471472 - `BetaManagedAgentsAlwaysAskPolicy object`473474 Tool calls require user confirmation before execution.475476 - `BetaManagedAgentsAutoPolicy object`477478 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.479480 - `BetaManagedAgentsMCPToolsetParams object`481482 Configuration for tools from an MCP server defined in `mcp_servers`.483484 - `type: "mcp_toolset"`485486 - `mcp_server_name: string`487488 Name of the MCP server. Must match a server name from the mcp_servers array. 1-255 characters.489490 minLength: 1, maxLength: 255491492 - `configs: optional array of BetaManagedAgentsMCPToolConfigParams`493494 Per-tool configuration overrides.495496 - `name: string`497498 Name of the MCP tool to configure. 1-128 characters.499500 minLength: 1, maxLength: 128501502 - `enabled: optional boolean or null`503504 Whether this tool is enabled. Overrides the `default_config` setting.505506 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`507508 Permission policy for tool execution.509510 - `BetaManagedAgentsAlwaysAllowPolicy object`511512 Tool calls are automatically approved without user confirmation.513514 - `BetaManagedAgentsAlwaysAskPolicy object`515516 Tool calls require user confirmation before execution.517518 - `BetaManagedAgentsAutoPolicy object`519520 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.521522 - `default_config: optional BetaManagedAgentsMCPToolsetDefaultConfigParams or null`523524 Default configuration for all tools from an MCP server.525526 - `enabled: optional boolean or null`527528 Whether tools are enabled by default. Defaults to true if not specified.529530 - `permission_policy: optional BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null`531532 Permission policy for tool execution.533534 - `BetaManagedAgentsAlwaysAllowPolicy object`535536 Tool calls are automatically approved without user confirmation.537538 - `BetaManagedAgentsAlwaysAskPolicy object`539540 Tool calls require user confirmation before execution.541542 - `BetaManagedAgentsAutoPolicy object`543544 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.545546 - `BetaManagedAgentsCustomToolParams object`547548 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.549550 - `type: "custom"`551552 - `description: string`553554 Description of what the tool does, shown to the agent to help it decide when to use the tool.555556 minLength: 1557558 - `input_schema: BetaManagedAgentsCustomToolInputSchema`559560 JSON Schema for custom tool input parameters.561562 - `type: "object"`563564 - `properties: optional map[unknown] or null`565566 - `required: optional array of string or null`567568 - `name: string`569570 Unique name for the tool. 1-128 characters; letters, digits, underscores, and hyphens.571572 minLength: 1, maxLength: 128573574- `budget: optional BetaManagedAgentsBudgetLimit or null`575576 A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches `max_list_cost`.577578 - `type: "limit"`579580 - `max_list_cost: BetaMonetaryAmount`581582 A monetary amount in a specific currency.583584 - `amount: string`585586 Amount in minor units of the currency, as an integer decimal string with no leading zeros: "2500" is $25.00 and "50" is fifty cents. A string rather than a number so no float rounding is ever applied.587588 - `currency: BetaCurrency`589590 Uppercase ISO-4217 currency code. `USD` is the only currency currently supported; the accepted set is closed and grows only when a new currency is priced.591592- `metadata: optional map[string] or null`593594 Metadata patch. Set a key to a string to upsert it, or to null to delete it. Omit the field to preserve.595596- `title: optional string or null`597598 Human-readable session title.599600 minLength: 1, maxLength: 500601602- `vault_ids: optional array of string`603604 Vault IDs (`vlt_*`) to attach to the session. Not yet supported; requests setting this field are rejected. Reserved for future use.605606## Returns607608- `BetaManagedAgentsSession object`609610 A Managed Agents `session`.611612 - `type: "session"`613614 - `id: string`615616 - `agent: BetaManagedAgentsSessionAgent`617618 Resolved `agent` definition for a `session`. Snapshot of the `agent` at `session` creation time.619620 - `type: "agent"`621622 - `id: string`623624 - `description: string or null`625626 - `mcp_servers: array of BetaManagedAgentsMCPServerURLDefinition`627628 - `type: "url"`629630 - `name: string`631632 - `url: string`633634 - `model: BetaManagedAgentsModelConfig`635636 Model identifier and configuration.637638 - `id: BetaManagedAgentsModel`639640 The model that will power your agent.641642 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.643644 - `"claude-fable-5-1" or "claude-sonnet-5" or "claude-fable-5" or 11 more`645646 The model that will power your agent.647648 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.649650 - `"claude-fable-5-1"`651652 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows653654 - `"claude-sonnet-5"`655656 High-performance model for coding and agents657658 - `"claude-fable-5"`659660 Next generation of intelligence for the hardest knowledge work and coding problems661662 - `"claude-opus-5"`663664 Powerful intelligence for long-running agents and coding665666 - `"claude-opus-4-8"`667668 Powerful intelligence for long-running agents and coding669670 - `"claude-opus-4-7"`671672 Powerful intelligence for long-running agents and coding673674 - `"claude-opus-4-6"`675676 Powerful intelligence for long-running agents and coding677678 - `"claude-sonnet-4-6"`679680 Best combination of speed and intelligence681682 - `"claude-haiku-4-5"`683684 Fastest model with near-frontier intelligence685686 - `"claude-haiku-4-5-20251001"`687688 Fastest model with near-frontier intelligence689690 - `"claude-opus-4-5"`691692 Powerful intelligence for long-running agents and coding693694 - `"claude-opus-4-5-20251101"`695696 Powerful intelligence for long-running agents and coding697698 - `"claude-sonnet-4-5"`699700 High-performance model for agents and coding701702 - `"claude-sonnet-4-5-20250929"`703704 High-performance model for agents and coding705706 - `string`707708 - `effort: optional BetaManagedAgentsEffortLow or BetaManagedAgentsEffortMedium or BetaManagedAgentsEffortHigh or 2 more`709710 How hard Claude works on each turn. Sets `output_config.effort` on every Messages call the session makes.711712 - `BetaManagedAgentsEffortLow object`713714 Low effort. Favors latency over reasoning depth.715716 - `type: "low"`717718 - `BetaManagedAgentsEffortMedium object`719720 Medium effort. Balances latency and reasoning depth.721722 - `type: "medium"`723724 - `BetaManagedAgentsEffortHigh object`725726 High effort. Favors reasoning depth.727728 - `type: "high"`729730 - `BetaManagedAgentsEffortXhigh object`731732 Extra-high effort. Not all models accept this level.733734 - `type: "xhigh"`735736 - `BetaManagedAgentsEffortMax object`737738 Maximum effort. Favors reasoning depth over latency.739740 - `type: "max"`741742 - `inference_geo: optional string`743744 Geographic region for model inference. When unset, requests fall through to the workspace's default_inference_geo.745746 - `speed: optional "standard" or "fast"`747748 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.749750 - `"standard"`751752 - `"fast"`753754 - `multiagent: BetaManagedAgentsSessionMultiagentCoordinator or null`755756 Resolved coordinator topology with full agent definitions for each roster member.757758 - `type: "coordinator"`759760 - `agents: array of BetaManagedAgentsSessionThreadAgent or BetaManagedAgentsAdvisor`761762 Full `agent` definitions the coordinator may spawn as session threads.763764 - `BetaManagedAgentsSessionThreadAgent object`765766 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`.767768 - `type: "agent"`769770 - `id: string`771772 - `description: string or null`773774 - `mcp_servers: array of BetaManagedAgentsMCPServerURLDefinition`775776 - `type: "url"`777778 - `name: string`779780 - `url: string`781782 - `model: BetaManagedAgentsModelConfig`783784 Model identifier and configuration.785786 - `name: string`787788 - `skills: array of BetaManagedAgentsAnthropicSkill or BetaManagedAgentsCustomSkill`789790 - `BetaManagedAgentsAnthropicSkill object`791792 A resolved Anthropic-managed skill.793794 - `type: "anthropic"`795796 - `skill_id: string`797798 - `version: string`799800 - `BetaManagedAgentsCustomSkill object`801802 A resolved user-created custom skill.803804 - `type: "custom"`805806 - `skill_id: string`807808 - `version: string`809810 - `system: string or null`811812 - `tools: array of BetaManagedAgentsAgentToolset20260401 or BetaManagedAgentsMCPToolset or BetaManagedAgentsCustomTool`813814 - `BetaManagedAgentsAgentToolset20260401 object`815816 - `type: "agent_toolset_20260401"`817818 - `configs: array of BetaManagedAgentsAgentToolConfig`819820 - `BetaManagedAgentsBashToolConfig object`821822 Configuration for the bash tool.823824 - `type: "bash"`825826 - `enabled: boolean`827828 - `name: "bash"`829830 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`831832 Permission policy for tool execution.833834 - `BetaManagedAgentsAlwaysAllowPolicy object`835836 Tool calls are automatically approved without user confirmation.837838 - `type: "always_allow"`839840 - `BetaManagedAgentsAlwaysAskPolicy object`841842 Tool calls require user confirmation before execution.843844 - `type: "always_ask"`845846 - `BetaManagedAgentsAutoPolicy object`847848 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.849850 - `type: "auto"`851852 - `BetaManagedAgentsEditToolConfig object`853854 Configuration for the edit tool.855856 - `type: "edit"`857858 - `enabled: boolean`859860 - `name: "edit"`861862 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`863864 Permission policy for tool execution.865866 - `BetaManagedAgentsAlwaysAllowPolicy object`867868 Tool calls are automatically approved without user confirmation.869870 - `BetaManagedAgentsAlwaysAskPolicy object`871872 Tool calls require user confirmation before execution.873874 - `BetaManagedAgentsAutoPolicy object`875876 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.877878 - `BetaManagedAgentsReadToolConfig object`879880 Configuration for the read tool.881882 - `type: "read"`883884 - `enabled: boolean`885886 - `name: "read"`887888 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`889890 Permission policy for tool execution.891892 - `BetaManagedAgentsAlwaysAllowPolicy object`893894 Tool calls are automatically approved without user confirmation.895896 - `BetaManagedAgentsAlwaysAskPolicy object`897898 Tool calls require user confirmation before execution.899900 - `BetaManagedAgentsAutoPolicy object`901902 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.903904 - `BetaManagedAgentsWriteToolConfig object`905906 Configuration for the write tool.907908 - `type: "write"`909910 - `enabled: boolean`911912 - `name: "write"`913914 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`915916 Permission policy for tool execution.917918 - `BetaManagedAgentsAlwaysAllowPolicy object`919920 Tool calls are automatically approved without user confirmation.921922 - `BetaManagedAgentsAlwaysAskPolicy object`923924 Tool calls require user confirmation before execution.925926 - `BetaManagedAgentsAutoPolicy object`927928 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.929930 - `BetaManagedAgentsGlobToolConfig object`931932 Configuration for the glob tool.933934 - `type: "glob"`935936 - `enabled: boolean`937938 - `name: "glob"`939940 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`941942 Permission policy for tool execution.943944 - `BetaManagedAgentsAlwaysAllowPolicy object`945946 Tool calls are automatically approved without user confirmation.947948 - `BetaManagedAgentsAlwaysAskPolicy object`949950 Tool calls require user confirmation before execution.951952 - `BetaManagedAgentsAutoPolicy object`953954 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.955956 - `BetaManagedAgentsGrepToolConfig object`957958 Configuration for the grep tool.959960 - `type: "grep"`961962 - `enabled: boolean`963964 - `name: "grep"`965966 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`967968 Permission policy for tool execution.969970 - `BetaManagedAgentsAlwaysAllowPolicy object`971972 Tool calls are automatically approved without user confirmation.973974 - `BetaManagedAgentsAlwaysAskPolicy object`975976 Tool calls require user confirmation before execution.977978 - `BetaManagedAgentsAutoPolicy object`979980 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.981982 - `BetaManagedAgentsWebFetchToolConfig object`983984 Configuration for the web_fetch tool.985986 - `type: "web_fetch"`987988 - `enabled: boolean`989990 - `name: "web_fetch"`991992 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`993994 Permission policy for tool execution.995996 - `BetaManagedAgentsAlwaysAllowPolicy object`997998 Tool calls are automatically approved without user confirmation.9991000 - `BetaManagedAgentsAlwaysAskPolicy object`10011002 Tool calls require user confirmation before execution.10031004 - `BetaManagedAgentsAutoPolicy object`10051006 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.10071008 - `allowed_domains: optional array of string`10091010 - `blocked_domains: optional array of string`10111012 - `max_content_tokens: optional number or null`10131014 format: int3210151016 - `BetaManagedAgentsWebSearchToolConfig object`10171018 Configuration for the web_search tool.10191020 - `type: "web_search"`10211022 - `enabled: boolean`10231024 - `name: "web_search"`10251026 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`10271028 Permission policy for tool execution.10291030 - `BetaManagedAgentsAlwaysAllowPolicy object`10311032 Tool calls are automatically approved without user confirmation.10331034 - `BetaManagedAgentsAlwaysAskPolicy object`10351036 Tool calls require user confirmation before execution.10371038 - `BetaManagedAgentsAutoPolicy object`10391040 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.10411042 - `allowed_domains: optional array of string`10431044 - `blocked_domains: optional array of string`10451046 - `user_location: optional BetaManagedAgentsUserLocation or null`10471048 Approximate user location for search result localization.10491050 - `type: "approximate"`10511052 Location precision. Only "approximate" is supported.10531054 - `city: optional string or null`10551056 City name.10571058 minLength: 1, maxLength: 25510591060 - `country: optional string or null`10611062 Two-letter ISO 3166-1 country code, uppercase.10631064 - `region: optional string or null`10651066 Region or state name.10671068 minLength: 1, maxLength: 25510691070 - `timezone: optional string or null`10711072 IANA timezone identifier, e.g. "America/Los_Angeles".10731074 minLength: 1, maxLength: 25510751076 - `default_config: BetaManagedAgentsAgentToolsetDefaultConfig`10771078 Resolved default configuration for agent tools.10791080 - `enabled: boolean`10811082 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`10831084 Permission policy for tool execution.10851086 - `BetaManagedAgentsAlwaysAllowPolicy object`10871088 Tool calls are automatically approved without user confirmation.10891090 - `BetaManagedAgentsAlwaysAskPolicy object`10911092 Tool calls require user confirmation before execution.10931094 - `BetaManagedAgentsAutoPolicy object`10951096 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.10971098 - `BetaManagedAgentsMCPToolset object`10991100 - `type: "mcp_toolset"`11011102 - `configs: array of BetaManagedAgentsMCPToolConfig`11031104 - `enabled: boolean`11051106 - `name: string`11071108 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`11091110 Permission policy for tool execution.11111112 - `BetaManagedAgentsAlwaysAllowPolicy object`11131114 Tool calls are automatically approved without user confirmation.11151116 - `BetaManagedAgentsAlwaysAskPolicy object`11171118 Tool calls require user confirmation before execution.11191120 - `BetaManagedAgentsAutoPolicy object`11211122 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.11231124 - `default_config: BetaManagedAgentsMCPToolsetDefaultConfig`11251126 Resolved default configuration for all tools from an MCP server.11271128 - `enabled: boolean`11291130 - `permission_policy: BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy`11311132 Permission policy for tool execution.11331134 - `BetaManagedAgentsAlwaysAllowPolicy object`11351136 Tool calls are automatically approved without user confirmation.11371138 - `BetaManagedAgentsAlwaysAskPolicy object`11391140 Tool calls require user confirmation before execution.11411142 - `BetaManagedAgentsAutoPolicy object`11431144 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.11451146 - `mcp_server_name: string`11471148 - `BetaManagedAgentsCustomTool object`11491150 A custom tool as returned in API responses.11511152 - `type: "custom"`11531154 - `description: string`11551156 - `input_schema: BetaManagedAgentsCustomToolInputSchema`11571158 JSON Schema for custom tool input parameters.11591160 - `type: "object"`11611162 - `properties: optional map[unknown] or null`11631164 - `required: optional array of string or null`11651166 - `name: string`11671168 - `version: number`11691170 format: int3211711172 - `BetaManagedAgentsAdvisor object`11731174 Platform advisor roster entry: a model the session's primary thread may consult mid-turn.11751176 - `type: "advisor"`11771178 - `model: string`11791180 The advisor model id.11811182 - `name: string`11831184 - `skills: array of BetaManagedAgentsAnthropicSkill or BetaManagedAgentsCustomSkill`11851186 - `BetaManagedAgentsAnthropicSkill object`11871188 A resolved Anthropic-managed skill.11891190 - `BetaManagedAgentsCustomSkill object`11911192 A resolved user-created custom skill.11931194 - `system: string or null`11951196 - `tools: array of BetaManagedAgentsAgentToolset20260401 or BetaManagedAgentsMCPToolset or BetaManagedAgentsCustomTool`11971198 - `BetaManagedAgentsAgentToolset20260401 object`11991200 - `BetaManagedAgentsMCPToolset object`12011202 - `BetaManagedAgentsCustomTool object`12031204 A custom tool as returned in API responses.12051206 - `version: number`12071208 format: int3212091210 - `archived_at: string or null`12111212 A timestamp in RFC 3339 format12131214 format: date-time12151216 - `budget: BetaManagedAgentsBudgetLimit or null`12171218 A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches `max_list_cost`.12191220 - `type: "limit"`12211222 - `max_list_cost: BetaMonetaryAmount`12231224 A monetary amount in a specific currency.12251226 - `amount: string`12271228 Amount in minor units of the currency, as an integer decimal string with no leading zeros: "2500" is $25.00 and "50" is fifty cents. A string rather than a number so no float rounding is ever applied.12291230 - `currency: BetaCurrency`12311232 Uppercase ISO-4217 currency code. `USD` is the only currency currently supported; the accepted set is closed and grows only when a new currency is priced.12331234 - `created_at: string`12351236 A timestamp in RFC 3339 format12371238 format: date-time12391240 - `environment_id: string`12411242 - `metadata: map[string]`12431244 - `outcome_evaluations: array of BetaManagedAgentsOutcomeEvaluationResource`12451246 Per-outcome evaluation state. One entry per `define_outcome` event sent to the session.12471248 - `type: "outcome_evaluation"`12491250 - `completed_at: string or null`12511252 A timestamp in RFC 3339 format12531254 format: date-time12551256 - `description: string`12571258 What the agent should produce.12591260 - `explanation: string or null`12611262 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.12631264 - `iteration: number`12651266 0-indexed revision cycle the outcome is currently on.12671268 format: int3212691270 - `outcome_id: string`12711272 Server-generated outc_ ID for this outcome.12731274 - `result: string`12751276 Current evaluation state. `pending` before the agent begins work; `running` while producing or revising; `evaluating` while the grader scores; `satisfied`/`max_iterations_reached`/`failed`/`interrupted` are terminal.12771278 - `resources: array of BetaManagedAgentsSessionResource`12791280 - `BetaManagedAgentsGitHubRepositoryResource object`12811282 - `type: "github_repository"`12831284 - `id: string`12851286 - `created_at: string`12871288 A timestamp in RFC 3339 format12891290 format: date-time12911292 - `mount_path: string`12931294 - `updated_at: string`12951296 A timestamp in RFC 3339 format12971298 format: date-time12991300 - `url: string`13011302 - `checkout: optional BetaManagedAgentsBranchCheckout or BetaManagedAgentsCommitCheckout or null`13031304 - `BetaManagedAgentsBranchCheckout object`13051306 - `type: "branch"`13071308 - `name: string`13091310 Branch name to check out.13111312 minLength: 1, maxLength: 25513131314 - `BetaManagedAgentsCommitCheckout object`13151316 - `type: "commit"`13171318 - `sha: string`13191320 Full commit SHA to check out.13211322 minLength: 7, maxLength: 6413231324 - `BetaManagedAgentsFileResource object`13251326 - `type: "file"`13271328 - `id: string`13291330 - `created_at: string`13311332 A timestamp in RFC 3339 format13331334 format: date-time13351336 - `file_id: string`13371338 - `mount_path: string`13391340 - `updated_at: string`13411342 A timestamp in RFC 3339 format13431344 format: date-time13451346 - `BetaManagedAgentsMemoryStoreResource object`13471348 A memory store attached to an agent session.13491350 - `type: "memory_store"`13511352 - `memory_store_id: string`13531354 The memory store ID (memstore_...). Must belong to the caller's organization and workspace.13551356 - `access: optional "read_write" or "read_only" or null`13571358 Access mode for an attached memory store.13591360 - `"read_write"`13611362 - `"read_only"`13631364 - `description: optional string`13651366 Description of the memory store, snapshotted at attach time. Rendered into the agent's system prompt. Empty string when the store has no description.13671368 - `instructions: optional string or null`13691370 Per-attachment guidance for the agent on how to use this store. Rendered into the memory section of the system prompt. Max 4096 chars.13711372 maxLength: 409613731374 - `mount_path: optional string or null`13751376 Filesystem path where the store is mounted in the session container, e.g. /mnt/memory/user-preferences. Derived from the store's name. Output-only.13771378 - `name: optional string or null`13791380 Display name of the memory store, snapshotted at attach time. Later edits to the store's name do not propagate to this resource.13811382 - `stats: BetaManagedAgentsSessionStats`13831384 Timing statistics for a session.13851386 - `active_seconds: optional number`13871388 Cumulative time in seconds the session spent in `running` status. Excludes idle time.13891390 format: double13911392 - `duration_seconds: optional number`13931394 Elapsed time since session creation in seconds. For terminated sessions, frozen at the final update.13951396 format: double13971398 - `status: "rescheduling" or "running" or "idle" or "terminated"`13991400 SessionStatus enum14011402 - `"rescheduling"`14031404 - `"running"`14051406 - `"idle"`14071408 - `"terminated"`14091410 - `title: string or null`14111412 - `updated_at: string`14131414 A timestamp in RFC 3339 format14151416 format: date-time14171418 - `usage: BetaManagedAgentsSessionUsage`14191420 Cumulative token usage for a session across all turns.14211422 - `active_seconds: optional number`14231424 Cumulative time in seconds during which the session had at least one thread in running status. Overlapping activity from concurrent threads is counted once, unlike `stats.active_seconds`, which sums each thread's own active time. This is the duration the session's runtime cost is priced on.14251426 format: double14271428 - `cache_creation: optional BetaManagedAgentsCacheCreationUsage`14291430 Prompt-cache creation token usage broken down by cache lifetime.14311432 - `ephemeral_1h_input_tokens: optional number`14331434 Tokens used to create 1-hour ephemeral cache entries.14351436 format: int3214371438 - `ephemeral_5m_input_tokens: optional number`14391440 Tokens used to create 5-minute ephemeral cache entries.14411442 format: int3214431444 - `cache_read_input_tokens: optional number`14451446 Total tokens read from prompt cache.14471448 format: int3214491450 - `input_tokens: optional number`14511452 Total input tokens consumed across all turns.14531454 format: int3214551456 - `list_cost: optional BetaMonetaryAmount or null`14571458 A monetary amount in a specific currency.14591460 - `output_tokens: optional number`14611462 Total output tokens generated across all turns.14631464 format: int3214651466 - `server_tool_use: optional BetaManagedAgentsServerToolUsage or null`14671468 Cumulative count of server-executed tool invocations, broken down by tool.14691470 - `web_fetch_requests: optional number`14711472 Number of server-executed web fetch requests.14731474 format: int3214751476 - `web_search_requests: optional number`14771478 Number of server-executed web search requests.14791480 format: int3214811482 - `vault_ids: array of string`14831484 Vault IDs attached to the session at creation. Empty when no vaults were supplied.14851486 - `deployment_id: optional string or null`14871488 Deployment ID when the session was created from a deployment reference. Null otherwise.14891490## Example14911492```bash1493curl https://api.anthropic.com/v1/sessions/$SESSION_ID \1494 -H 'Content-Type: application/json' \1495 -H 'anthropic-version: 2023-06-01' \1496 -H 'anthropic-beta: managed-agents-2026-04-01' \1497 -H "X-Api-Key: $ANTHROPIC_API_KEY" \1498 -d '{1499 "title": "Order #1234 inquiry"1500 }'1501```15021503### Response (200)15041505```json1506{1507 "id": "sesn_011CZkZAtmR3yMPDzynEDxu7",1508 "agent": {1509 "id": "agent_011CZkYpogX7uDKUyvBTophP",1510 "description": "A general-purpose starter agent.",1511 "mcp_servers": [1512 {1513 "name": "example-mcp",1514 "type": "url",1515 "url": "https://example-server.modelcontextprotocol.io/sse"1516 }1517 ],1518 "model": {1519 "id": "claude-opus-5",1520 "effort": {1521 "type": "low"1522 },1523 "inference_geo": "inference_geo",1524 "speed": "standard"1525 },1526 "multiagent": {1527 "agents": [1528 {1529 "id": "agent_011CZkYqphY8vELVzwCUpqiQ",1530 "description": "A focused research subagent.",1531 "mcp_servers": [1532 {1533 "name": "example-mcp",1534 "type": "url",1535 "url": "https://example-server.modelcontextprotocol.io/sse"1536 }1537 ],1538 "model": {1539 "id": "claude-opus-5",1540 "effort": {1541 "type": "low"1542 },1543 "inference_geo": "inference_geo",1544 "speed": "standard"1545 },1546 "name": "Researcher",1547 "skills": [1548 {1549 "skill_id": "xlsx",1550 "type": "anthropic",1551 "version": "1"1552 }1553 ],1554 "system": "You are a research subagent that gathers and summarises sources for the coordinating agent.",1555 "tools": [1556 {1557 "configs": [1558 {1559 "enabled": true,1560 "name": "bash",1561 "permission_policy": {1562 "type": "always_allow"1563 },1564 "type": "bash"1565 }1566 ],1567 "default_config": {1568 "enabled": true,1569 "permission_policy": {1570 "type": "always_ask"1571 }1572 },1573 "type": "agent_toolset_20260401"1574 }1575 ],1576 "type": "agent",1577 "version": 11578 }1579 ],1580 "type": "coordinator"1581 },1582 "name": "My First Agent",1583 "skills": [1584 {1585 "skill_id": "xlsx",1586 "type": "anthropic",1587 "version": "1"1588 },1589 {1590 "skill_id": "skill_011CZkZFNu9hAbo3jZPRgTlx",1591 "type": "custom",1592 "version": "2"1593 }1594 ],1595 "system": "You are a general-purpose agent that can research, write code, run commands, and use connected tools to complete the user's task end to end.",1596 "tools": [1597 {1598 "configs": [1599 {1600 "enabled": true,1601 "name": "bash",1602 "permission_policy": {1603 "type": "always_allow"1604 },1605 "type": "bash"1606 }1607 ],1608 "default_config": {1609 "enabled": true,1610 "permission_policy": {1611 "type": "always_ask"1612 }1613 },1614 "type": "agent_toolset_20260401"1615 }1616 ],1617 "type": "agent",1618 "version": 11619 },1620 "archived_at": null,1621 "budget": {1622 "max_list_cost": {1623 "amount": "2500",1624 "currency": "USD"1625 },1626 "type": "limit"1627 },1628 "created_at": "2026-03-15T10:00:00Z",1629 "environment_id": "env_011CZkZ9X2dpNyB7HsEFoRfW",1630 "metadata": {},1631 "outcome_evaluations": [1632 {1633 "completed_at": "2026-03-15T10:02:31Z",1634 "description": "Produce a 2-page summary as summary.md",1635 "explanation": "All five sections present with inline citations.",1636 "iteration": 0,1637 "outcome_id": "outc_011CZkZRSw2kEfs6ncTVljxP",1638 "result": "satisfied",1639 "type": "outcome_evaluation"1640 }1641 ],1642 "resources": [1643 {1644 "id": "sesrsc_011CZkZBJq5dWxk9fVLNcPht",1645 "created_at": "2026-03-15T10:00:00Z",1646 "file_id": "file_011CNha8iCJcU1wXNR6q4V8w",1647 "mount_path": "/uploads/receipt.pdf",1648 "type": "file",1649 "updated_at": "2026-03-15T10:00:00Z"1650 },1651 {1652 "id": "sesrsc_011CZkZCKr6eXyl0gWMOdQiu",1653 "created_at": "2026-03-15T10:00:00Z",1654 "mount_path": "/workspace/example-repo",1655 "type": "github_repository",1656 "updated_at": "2026-03-15T10:00:00Z",1657 "url": "https://github.com/example-org/example-repo",1658 "checkout": {1659 "name": "main",1660 "type": "branch"1661 }1662 }1663 ],1664 "stats": {1665 "active_seconds": 0,1666 "duration_seconds": 01667 },1668 "status": "idle",1669 "title": "Order #1234 inquiry",1670 "type": "session",1671 "updated_at": "2026-03-15T10:00:00Z",1672 "usage": {1673 "active_seconds": 0,1674 "cache_creation": {1675 "ephemeral_1h_input_tokens": 0,1676 "ephemeral_5m_input_tokens": 01677 },1678 "cache_read_input_tokens": 0,1679 "input_tokens": 0,1680 "list_cost": {1681 "amount": "2500",1682 "currency": "USD"1683 },1684 "output_tokens": 0,1685 "server_tool_use": {1686 "web_fetch_requests": 0,1687 "web_search_requests": 31688 }1689 },1690 "vault_ids": [1691 "vlt_011CZkZDLs7fYzm1hXNPeRjv"1692 ],1693 "deployment_id": "deployment_id"1694}1695```1696