Unverified50% confidenceSolutionExact time
当Skill数量达数百上千时，应采用召回加重排方案：先用embedding或关键词检索召回Top 10候选，再交给大模型做最终裁决
1
Sources
50%
Confidence
Long-term
Relevance
7/20/2026
First Seen
Sources
Related Claims
Unverified当前工具选择策略高度依赖模型的即时判断，缺乏系统性优化，在企业场景中可用工具可达数十乃至数百个，模型选择时易出现幻觉、重复调用或选错工具72% similarUnverified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67% similarUnverifiedAssigning each agent a dedicated capability boundary and prompt often outperforms having a single model handle everything, a principle that is connected to the effectiveness of chain-of-thought prompting.65% similarUnverifiedSkill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65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/566482API
curl https://kongchang.com/api/v1/knowledge/claims/566482MCP
get_claim(id=566482)