Reflection Agent

A reflection agent is an AI agent that critiques and revises its own outputs before returning a final result.

After completing an initial draft or action plan, the agent re-evaluates it against quality criteria or constraints and iterates until the output meets the required standard. Reflection improves accuracy and reduces hallucination but increases token consumption and latency.

Closely related concepts include Quality Gate (AI), Evaluation Harness (Agent), Reflection, Ground Truth Evaluation. Each is defined in the Prefactor glossary.