Skip to main content

talonic_submit_agent_task

Return the result of a claimed Agent task. Output keys must match output_contract exactly, required fields must be present, and every value must match its declared data type. The platform validates the entire result before writing any cell.

ParameterTypeDescription
task_id *UUIDClaimed task ID.
execution_epoch *integerExact epoch from the current claim.
outputs *objectDeclared field key to `{ value, confidence?, reasoning? }`. Confidence is 0–1; reasoning is at most 4,000 characters.
summarystringOptional overall result summary, at most 4,000 characters.
Submit result
{
  "task_id": "11111111-1111-4111-8111-111111111111",
  "execution_epoch": 3,
  "outputs": {
    "risk_score": {
      "value": 0.72,
      "confidence": 0.88,
      "reasoning": "Two adverse indicators in the supplied evidence."
    }
  },
  "summary": "Supplier requires enhanced review."
}
Successful outputs are attributed to AI Agent (MCP) in the audit trail and stored with mcp_agent provenance. They do not receive human-review authority in v1.

Frequently asked questions

Can an Agent task return extra fields it discovered?+
No. Undeclared fields and invalid types are rejected before any value is written. Add desired outputs to the Agent stage contract first, then run a new document task.