Sesi
Models
beta_managed_agents_advisor_params: object{ model, type }Platform advisor roster entry: a model the session's primary thread may consult mid-turn. At most one per roster; the entry occupies the roster name anthropic.advisor.
Platform advisor roster entry: a model the session's primary thread may consult mid-turn. At most one per roster; the entry occupies the roster name anthropic.advisor.
model: stringA Claude model id. The model must be permitted as an advisor for this agent's model — see the sessions/threads/advisor spec.
A Claude model id. The model must be permitted as an advisor for this agent's model — see the sessions/threads/advisor spec.
beta_managed_agents_agent_message_preview: object{ id, type }
The id the buffered agent.message will carry if it is emitted. Matches the event_id on this preview's event_delta events.
beta_managed_agents_agent_params: object{ id, type, version }Specification for an Agent. Provide a specific version or use the short-form agent="agent_id" for the most recent version
Specification for an Agent. Provide a specific version or use the short-form agent="agent_id" for the most recent version
id: stringThe agent ID.
The agent ID.
version: optional numberThe specific agent version to use. Omit to use the latest version. Must be at least 1 if specified.
The specific agent version to use. Omit to use the latest version. Must be at least 1 if specified.
beta_managed_agents_agent_thinking_preview: object{ id, type }
The id the buffered agent.thinking will carry if it is emitted. Start-only — no event_delta events follow.
beta_managed_agents_agent_with_overrides_params: object{ id, type, mcp_servers, 5 more }Reference to an agent plus optional configuration overrides. Each provided field replaces the agent's value for the caller's use; the agent resource is unchanged.
Reference to an agent plus optional configuration overrides. Each provided field replaces the agent's value for the caller's use; the agent resource is unchanged.
beta_managed_agents_branch_checkout: object{ name, type }
name: stringBranch name to check out.
Branch name to check out.
beta_managed_agents_budget_limit: object{ max_list_cost, type }A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches max_list_cost.
A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches max_list_cost.
max_list_cost: object{ amount, currency }A monetary amount in a specific currency.
A monetary amount in a specific currency.
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.
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.
beta_managed_agents_cache_creation_usage: object{ ephemeral_1h_input_tokens, ephemeral_5m_input_tokens }Prompt-cache creation token usage broken down by cache lifetime.
Prompt-cache creation token usage broken down by cache lifetime.
ephemeral_1h_input_tokens: optional numberTokens used to create 1-hour ephemeral cache entries.
Tokens used to create 1-hour ephemeral cache entries.
ephemeral_5m_input_tokens: optional numberTokens used to create 5-minute ephemeral cache entries.
Tokens used to create 5-minute ephemeral cache entries.
beta_managed_agents_commit_checkout: object{ sha, type }
sha: stringFull commit SHA to check out.
Full commit SHA to check out.
beta_managed_agents_deleted_session: object{ id, type }Confirmation that a session has been permanently deleted.
Confirmation that a session has been permanently deleted.
beta_managed_agents_delta_content: object{ content, type, index }
content: object{ text, type }Regular text content.
Regular text content.
text: stringThe text content.
The text content.
index: optional numberWhich entry in the previewed event's content array this fragment lands in. Insert content as that entry when the index is new; append to the existing entry otherwise.
Which entry in the previewed event's content array this fragment lands in. Insert content as that entry when the index is new; append to the existing entry otherwise.
beta_managed_agents_delta_event: object{ delta, event_id, type }An incremental update to an event that is still being streamed. Deltas are best-effort and may stop early; when the buffered event with id == event_id is produced it carries the complete content. A model request that ends early (an error or interrupt) produces no buffered event — its terminal span.model_request_end closes the preview. Only sent on stream connections that opt in via event_deltas; never appears in event history.
An incremental update to an event that is still being streamed. Deltas are best-effort and may stop early; when the buffered event with id == event_id is produced it carries the complete content. A model request that ends early (an error or interrupt) produces no buffered event — its terminal span.model_request_end closes the preview. Only sent on stream connections that opt in via event_deltas; never appears in event history.
beta_managed_agents_delta_type: "agent.message" or "agent.thinking"EventDeltaType enum
EventDeltaType enum
beta_managed_agents_file_resource_params: object{ file_id, type, mount_path }Mount a file uploaded via the Files API into the session.
Mount a file uploaded via the Files API into the session.
file_id: stringID of a previously uploaded file.
ID of a previously uploaded file.
mount_path: optional stringMount path in the container. Defaults to /mnt/session/uploads/<file_id>.
Mount path in the container. Defaults to /mnt/session/uploads/<file_id>.
beta_managed_agents_github_repository_resource_params: object{ authorization_token, type, url, 2 more }Mount a GitHub repository into the session's container.
Mount a GitHub repository into the session's container.
beta_managed_agents_memory_store_resource_param: object{ memory_store_id, type, access, instructions }Parameters for attaching a memory store to an agent session.
Parameters for attaching a memory store to an agent session.
The memory store ID (memstore_...). Must belong to the caller's organization and workspace.
access: optional "read_write" or "read_only"Access mode for an attached memory store.
Access mode for an attached memory store.
instructions: optional stringPer-attachment guidance for the agent on how to use this store. Rendered into the memory section of the system prompt. Max 4096 chars.
Per-attachment guidance for the agent on how to use this store. Rendered into the memory section of the system prompt. Max 4096 chars.
beta_managed_agents_multiagent: object{ agents, type }Resolved coordinator topology with a concrete agent roster.
Resolved coordinator topology with a concrete agent roster.
beta_managed_agents_multiagent_params: object{ agents, type }A coordinator topology: the session's primary thread orchestrates work by spawning session threads, each running an agent drawn from the agents roster.
A coordinator topology: the session's primary thread orchestrates work by spawning session threads, each running an agent drawn from the agents roster.
beta_managed_agents_multiagent_roster_entry_params: string or BetaManagedAgentsAgentParams { id, type, version } or BetaManagedAgentsMultiagentSelfParams { type } or BetaManagedAgentsAdvisorParams { model, type }An entry in a multiagent roster: an agent ID string, a versioned agent reference, or self.
An entry in a multiagent roster: an agent ID string, a versioned agent reference, or self.
beta_managed_agents_outcome_evaluation_resource: object{ completed_at, description, explanation, 4 more }Evaluation state for a single outcome defined via a define_outcome event.
Evaluation state for a single outcome defined via a define_outcome event.
completed_at: stringA timestamp in RFC 3339 format
A timestamp in RFC 3339 format
What the agent should produce.
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.
iteration: number0-indexed revision cycle the outcome is currently on.
0-indexed revision cycle the outcome is currently on.
Server-generated outc_ ID for this outcome.
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.
beta_managed_agents_server_tool_usage: object{ web_fetch_requests, web_search_requests }Cumulative count of server-executed tool invocations, broken down by tool.
Cumulative count of server-executed tool invocations, broken down by tool.
web_fetch_requests: optional numberNumber of server-executed web fetch requests.
Number of server-executed web fetch requests.
web_search_requests: optional numberNumber of server-executed web search requests.
Number of server-executed web search requests.
beta_managed_agents_session: object{ id, agent, archived_at, 14 more }A Managed Agents session.
A Managed Agents session.
beta_managed_agents_session_agent: object{ id, description, mcp_servers, 8 more }Resolved agent definition for a session. Snapshot of the agent at session creation time.
Resolved agent definition for a session. Snapshot of the agent at session creation time.
beta_managed_agents_session_agent_update: object{ mcp_servers, tools }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.
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.
beta_managed_agents_session_multiagent_coordinator: object{ agents, type }Resolved coordinator topology with full agent definitions for each roster member.
Resolved coordinator topology with full agent definitions for each roster member.
beta_managed_agents_session_stats: object{ active_seconds, duration_seconds }Timing statistics for a session.
Timing statistics for a session.
active_seconds: optional numberCumulative time in seconds the session spent in running status. Excludes idle time.
Cumulative time in seconds the session spent in running status. Excludes idle time.
duration_seconds: optional numberElapsed time since session creation in seconds. For terminated sessions, frozen at the final update.
Elapsed time since session creation in seconds. For terminated sessions, frozen at the final update.
beta_managed_agents_session_updated_event: object{ id, processed_at, type, 4 more }Emitted when an UpdateSession request changed at least one field. Carries only the fields that changed; absent fields were not part of the update. The new configuration applies from the next turn.
Emitted when an UpdateSession request changed at least one field. Carries only the fields that changed; absent fields were not part of the update. The new configuration applies from the next turn.
beta_managed_agents_session_usage: object{ active_seconds, cache_creation, cache_read_input_tokens, 4 more }Cumulative token usage for a session across all turns.
Cumulative token usage for a session across all turns.
beta_managed_agents_session_usage_event: object{ id, processed_at, type, 2 more }Periodic snapshot of the session's cumulative usage and tracked list cost.
Periodic snapshot of the session's cumulative usage and tracked list cost.
beta_managed_agents_start_event: object{ event, type }Opens a preview of a buffered event. Carries the previewed event's type and id only. Followed by zero or more event_delta events with the same event id, normally concluded by the buffered event carrying that id. If the producing model request ends without that event (an error or interrupt mid-stream), its terminal span.model_request_end closes the preview. Only sent on stream connections that opt in via event_deltas; never appears in event history.
Opens a preview of a buffered event. Carries the previewed event's type and id only. Followed by zero or more event_delta events with the same event id, normally concluded by the buffered event carrying that id. If the producing model request ends without that event (an error or interrupt mid-stream), its terminal span.model_request_end closes the preview. Only sent on stream connections that opt in via event_deltas; never appears in event history.
beta_managed_agents_start_event_preview: BetaManagedAgentsAgentMessagePreview { id, type } or BetaManagedAgentsAgentThinkingPreview { id, type }
beta_managed_agents_agent_message_preview: object{ id, type }
The id the buffered agent.message will carry if it is emitted. Matches the event_id on this preview's event_delta events.
beta_managed_agents_agent_thinking_preview: object{ id, type }
The id the buffered agent.thinking will carry if it is emitted. Start-only — no event_delta events follow.
beta_managed_agents_system_content_block: object{ text, type }Regular text content.
Regular text content.
text: stringThe text content.
The text content.
beta_managed_agents_system_message_event: object{ id, content, type, processed_at }A mid-conversation system message event. Carries system-role content that is appended to the session as a role: "system" turn.
A mid-conversation system message event. Carries system-role content that is appended to the session as a role: "system" turn.
Unique identifier for this event.
content: array of BetaManagedAgentsSystemContentBlock { text, type }System content blocks. Text-only.
System content blocks. Text-only.
text: stringThe text content.
The text content.
processed_at: optional stringA timestamp in RFC 3339 format
A timestamp in RFC 3339 format
beta_managed_agents_user_tool_result_event: object{ id, tool_use_id, type, 4 more }Event sent by the client providing the result of an agent-toolset tool execution. Only valid on self_hosted environments, where sandbox-routed tools are executed by the client rather than the server.
Event sent by the client providing the result of an agent-toolset tool execution. Only valid on self_hosted environments, where sandbox-routed tools are executed by the client rather than the server.