Skip to main content

agrag.eval.trajectory.task_completion

task_completion(judge:DeepEvalBaseLLM, *, threshold:float = 0.5) -> BaseMetric

Build the judged metric for task completion.

Scores whether the run achieved the question's goal, from the question, the answer and the tool calls, with one judge call. The gate is the mean over questions, following the answer-quality eval, so no median of repeated calls is needed.

Parameters:

  • judge (DeepEvalBaseLLM) – The judge model.
  • threshold (float) – The minimum score that counts as success.

Returns:

  • BaseMetric – The task completion metric.