Concept批评者代理 / Critic代理
Critic Agent
A role in multi-agent systems dedicated to reviewing and challenging the outputs of other agents. By introducing an independent perspective, the critic agent identifies design flaws and logical gaps, serving as an adversarial verification mechanism for improving the quality of AI system outputs.
Timeline (last 90 days)
Sep 11
Introducing a Critic Agent to challenge the first agent's design is an option, but two AIs may share the same blind spots, making it more suitable as a supplementary measure.
Unverified50%