Unverified50% confidenceTradeoffExact time
每一个可被自动调用的技能都会占用上下文窗口，技能越多，模型决策时需要处理的描述信息越多，增加token消耗并可能导致错误的调用判断
1
Sources
50%
Confidence
Long-term
Relevance
8/12/2026
First Seen
Sources
Related Claims
UnverifiedSkills采用渐进式披露设计，平时只向模型提供技能的名称和描述，具体执行规则在需要时才加载78% similarUnverifiedWhat truly differentiates the three skills is automation reusability—namely, whether the judgment exercised during interaction can be written into reusable code78% similarUnverified研究表明，相比单一的全能提示词，为不同任务赋予专门角色和上下文的智能体能够显著提升输出质量76% similarVerifiedSkills enable on-demand loading of specialized knowledge through a progressive disclosure mechanism, loading a particular type of expertise into context only when the task genuinely requires it.76% similarUnverifiedSkills involve encapsulating a capability to be automatically discovered and used by the model, whereas Workflow requires explicit triggering with the ultrawork keyword.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/734995API
curl https://kongchang.com/api/v1/knowledge/claims/734995MCP
get_claim(id=734995)