Create Agent Log
Create an agent log entry.
Called by agent runners to report their actions. The log entry is persisted to SurrealDB and published to NATS for real-time streaming to connected mission control clients.
Returns: Created agent log entry with server-populated fields.
Documentation Index
Fetch the complete documentation index at: https://docs.pragmatiks.io/llms.txt
Use this file to discover all available pages before exploring further.
Body
Request body for creating an agent log entry.
Attributes: log_type: Category of this log entry. message: Human-readable description of what happened. instance_id: Agent instance that produced this log entry. metadata: Flexible payload (tool name, arguments, error details, etc.).
Types of agent log entries.
trigger_fired, agent_started, agent_completed, agent_error, tool_call, tool_result, reasoning, decision, plan_created, plan_updated, step_started, step_completed Response
Successful Response
A recorded action or decision by an AI agent.
Agent logs live inside each organization's SurrealDB namespace —
the namespace is the organization boundary, so no per-record
organization_id field is carried on the row.
Attributes: id: SurrealDB record ID (populated by database). agent_resource_id: SurrealDB ID of the agent resource. instance_id: Agent instance that produced this log entry. log_type: Category of this log entry. message: Human-readable description of what happened. metadata: Flexible payload (tool name, arguments, error details, etc.). timestamp: When this action occurred. expires_at: When this log entry should be garbage collected.
Types of agent log entries.
trigger_fired, agent_started, agent_completed, agent_error, tool_call, tool_result, reasoning, decision, plan_created, plan_updated, step_started, step_completed