{
"name": "<string>",
"session_id": "<string>",
"success": true,
"duration_ms": 123,
"provider": "<string>",
"model": "<string>",
"complete": true,
"escalate": true,
"result": "<unknown>",
"thinking": "<string>",
"error": {
"code": "AGENT_TIMEOUT",
"message": "Agent execution exceeded timeout limit",
"details": "<unknown>",
"retryable": false
},
"warnings": [
"<string>"
],
"input_tokens": 123,
"output_tokens": 123,
"total_tokens": 123,
"cost": 123,
"summarization_status": "none",
"message_count": 123,
"token_usage": 123,
"instruction_metadata": {
"instructions_processed": true,
"instruction_count": 123,
"was_refined": true,
"processing_time_ms": 123
},
"ai_summary_metadata": {
"summarization_performed": true,
"model": "<string>",
"original_tokens": 123,
"summary_tokens": 123,
"compression_ratio": 123,
"messages_processed": 123,
"processing_time_ms": 123,
"fallback_used": true
}
}Invoke an AI assistant
Invokes an AI assistant with specified configuration and prompt. Use this to execute assistants with inline definitions or reference pre-configured assistants by assistant subject.
{
"name": "<string>",
"session_id": "<string>",
"success": true,
"duration_ms": 123,
"provider": "<string>",
"model": "<string>",
"complete": true,
"escalate": true,
"result": "<unknown>",
"thinking": "<string>",
"error": {
"code": "AGENT_TIMEOUT",
"message": "Agent execution exceeded timeout limit",
"details": "<unknown>",
"retryable": false
},
"warnings": [
"<string>"
],
"input_tokens": 123,
"output_tokens": 123,
"total_tokens": 123,
"cost": 123,
"summarization_status": "none",
"message_count": 123,
"token_usage": 123,
"instruction_metadata": {
"instructions_processed": true,
"instruction_count": 123,
"was_refined": true,
"processing_time_ms": 123
},
"ai_summary_metadata": {
"summarization_performed": true,
"model": "<string>",
"original_tokens": 123,
"summary_tokens": 123,
"compression_ratio": 123,
"messages_processed": 123,
"processing_time_ms": 123,
"fallback_used": true
}
}Authorizations
JWT Authorization header using the Bearer scheme.
Query Parameters
Cancel identifier for this assistant invocation. NOTE: this endpoint does NOT read this query param (nor the request-body cancel_token) — to cancel a running invocation, call POST /hub/agent/cancel instead. Kept for compatibility.
Body
Provide either subject (an existing assistant) OR an inline agent definition — 400 if neither. prompt (string or object) is expected but not strictly enforced by the engine. The old example field agent_subject is NOT read — use subject.
Inline assistant definition - use this to define assistant configuration directly in the request.
Show child attributes
Show child attributes
Subject of assistant assistant to invoke - use this to reference a pre-configured assistant.
"ms.hub.config.agent.8f5568aa-c88c-32d7-adfb-8565d367fb24"
The input prompt or query for the assistant to process - can be string or structured object.
"Analyze sales trends"
Session identifier for conversation continuity - omit for one-shot interactions.
"session_abc123"
NOTE: not read on invoke. To cancel a running invocation, call POST /hub/agent/cancel instead.
"my-session-token-123"
Visibility level for the invocation (user: private, team: shared).
user, team Additional knowledge sources for this specific invocation using URI schemes (kv://, obj://, str://, sid://, dta://).
Subject for async response publishing.
Whether to wait for synchronous response (true) or return immediately (false).
Optional folder/space path to organise the resulting session.
Override the provider for this invocation (requires model).
claude, anthropic Override the model for this invocation (requires llm_provider).
If true, persist the message but do not run the assistant.
Response
Assistant invoked successfully.
Assistant name from the request.
Session ID for conversation continuity.
Whether the invocation succeeded.
Total execution time in milliseconds.
LLM provider used for this invocation.
Model identifier used for this invocation.
Whether the assistant has completed its task (true) or requires more interaction (false).
Whether the assistant is requesting human intervention or admin escalation.
Assistant response - structured according to OutputSchema if defined, otherwise free-form string.
Model's reasoning or thought process extracted from the response.
Error details if success is false.
Show child attributes
Show child attributes
Non-fatal warnings from provisioning or execution.
Number of input tokens consumed.
Number of output tokens generated.
Total tokens used (input + output).
Estimated cost in USD for this invocation.
Status of conversation summarization.
none, pending, completed Total number of messages in the session conversation history.
Current conversation context usage in tokens.
Metadata about instruction processing if instructions were provided.
Show child attributes
Show child attributes
Metadata about AI summarization if performed.
Show child attributes
Show child attributes