Unverified80% confidenceOpinionTime unknown
Skill的正确使用方式是渐进式优化：使用→发现问题→加规则→再测试，而非追求一次完美
1
Sources
80%
Confidence
Long-term
Relevance
6/2/2026
First Seen
Sources
Related Entities
Related Claims
Unverified当 Skill 数量膨胀到一定程度，Skill 路由从简单的选择问题演变为检索问题79% similarUnverified确定性优化可能以牺牲模型精度或表达能力为代价，需在质量与能效之间寻找平衡点76% similarUnverifiedUnlike supervised learning, which relies on labeled data, reinforcement learning does not require humans to provide correct answers; instead, it allows agents to autonomously discover optimal behavioral paths through trial and error.75% similarUnverifiedSkill分为能力型Skill和偏好型Skill，能力型随模型变强会被淘汰，偏好型更持久需要用评估保护防止性能倒退73% similarUnverifiedAdvanced agentic harnesses have built-in retry logic, parameter validation, and self-correction mechanisms, enabling agents to understand error messages and adjust strategies when encountering failures.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/37435API
curl https://kongchang.com/api/v1/knowledge/claims/37435MCP
get_claim(id=37435)