{"id":"febfcc2c-e8f2-4dc3-8f4e-80bd555a488b","slug":"clawhub-hauxu-self-correction","name":"self-correction","description":"模型自我检查与纠错技能。当用户表达反对意见或指出错误时触发，如\"不对\"、\"你回答错了\"、\"不是这样\"、\"重新理解\"、\"等等\"、\"我说的不是这个意思\"等表达。用于重新分析和优化用户原始意图，而非修改上一轮回答。","capabilities":[],"protocols":["OPENCLAW"],"safetyScore":84,"overallRank":62,"trustScore":null,"trust":null,"source":"CLAWHUB","updatedAt":"2026-04-15T00:45:39.800Z"}