Skip to main content
Version: latest

agrag.agents.result.AgentRunResult

Bases: TypedDict

Outcome, answer, evidence references and conversation for one run.

Attributes:

  • answer (str) – Final answer text.
  • status (Literal['answered', 'insufficient_evidence', 'partial_retrieval', 'partial', 'recording_failed', 'budget_exhausted']) – Outcome after evidence, write and budget checks.
  • citations (list[FetchOutcome]) – Complete fetched evidence supporting the answer.
  • invalid_citations (list[str]) – Cited evidence ids that did not pass validation.
  • recording_failures (list[str]) – Descriptions of task writes that did not complete.
  • task (AgentTaskRef) – Task name and latest persisted event sequence.
  • messages (list[Any]) – Messages produced by the agent.

answer​

answer: str

citations​

citations: list[FetchOutcome]

invalid_citations​

invalid_citations: list[str]

messages​

messages: list[Any]

recording_failures​

recording_failures: list[str]

status​

status: Literal['answered', 'insufficient_evidence', 'partial_retrieval', 'partial', 'recording_failed', 'budget_exhausted']

task​

task: AgentTaskRef