For AI agents: a documentation index is available at the root level at /llms.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Get a single execution run for a Task Agent by run ID.
Returns the same run metadata shape as the run listing endpoint, including
status, token usage, files, tool invocation summaries, phase timings, fallback
metadata, model usage rows, and execution identity when available.
Authentication
AuthorizationBearer
API key or access token, sent as Authorization: Bearer <token>. The simplest option: create an API key in your dashboard under Settings, then paste it here.
OR
AuthorizationBearer
API key or access token, sent as Authorization: Bearer <token>. The simplest option: create an API key in your dashboard under Settings, then paste it here.
Path parameters
org_idstringRequiredformat: "uuid"
The organization ID
project_idstringRequiredformat: "uuid"
The project ID
task_agent_idstringRequiredformat: "uuid"
The task agent ID
run_idstringRequiredformat: "uuid"
The task agent run ID
Response
Task Agent run
idstringformat: "uuid"
task_agent_idstringformat: "uuid"
session_idstring
ADK session ID (string format).
user_idstring or nullformat: "uuid"
input_tokensinteger
output_tokensinteger
total_tokensinteger
charged_creditsstring or nullformat: "^(?!^[-+.]*$)[+-]?0*\d*\.?\d*$"
createdstring
completedstring or null
duration_msinteger or null
statusenum
Status of a TaskAgent run.
Allowed values:
is_errorboolean
task_tools_invokedlist of objects or null
is_chaos_runbooleanRead-only
task_agent_revision_idstring or nullOptionalformat: "uuid"
adk_session_user_idstring or nullOptional
actor_user_idstring or nullOptionalformat: "uuid"
User whose authority caused this run.
actor_basisstring or nullOptional
Reason the run actor was trusted.
execution_surfaceenum or nullOptional
Platform surface through which execution entered.
execution_source_idstring or nullOptional
Surface-specific source identifier for correlating this run.
invocation_idstring or nullOptional
invocation_idslist of strings or nullOptional
is_continuationbooleanOptionalDefaults to false
primary_llm_model_idstring or nullOptionalformat: "uuid"
fallback_llm_model_idstring or nullOptionalformat: "uuid"
used_fallback_modelbooleanOptionalDefaults to false
primary_failure_reasonstring or nullOptional
total_charged_creditsstring or nullOptionalformat: "^(?!^[-+.]*$)[+-]?0*\d*\.?\d*$"
approval_request_idslist of stringsOptional
IDs of approval requests linked to this TaskAgentRun.
approval_requestslist of objectsOptional
Compact approval request summaries linked to this TaskAgentRun.
had_task_tool_failbooleanOptionalDefaults to false
had_mcp_tool_failbooleanOptionalDefaults to false
had_integration_tool_failbooleanOptionalDefaults to false
skills_activatedlist of objects or nullOptional
mcp_tools_invokedlist of objects or nullOptional
integration_tools_invokedlist of objects or nullOptional
phase_timingslist of objects or nullOptional
context_usageobject or nullOptional
chaos_configobject or nullOptional
Run-scoped TaskAgent chaos policy.
chaos_eventslist of objects or nullOptional
termination_reasonstring or nullOptional
last_activity_atstring or nullOptional
trigger_typestringOptionalDefaults to manual
How the run was triggered. Non-trigger sources include manual, api, and slack; trigger executions use their trigger type value.
model_usageslist of objectsOptional
fileslist of objectsOptionalDefaults to []
Files associated with this agent run (both uploaded and generated by task tools)
