Unverified50% confidenceFactExact time
面向客户的Agent只能依赖模型自主触发（model-invoked）的Skills，而无法依赖用户主动调用
1
Sources
50%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
Related Claims
Unverified接入外部 Agent 后用户仍可按任务需求自由选择模型与推理强度76% similarUnverifiedAgent Skill adopts a lazy-loading mechanism whereby the Agent first reads only the metadata (name and description) of each Skill and loads the full instructions only after matching the user's need, so having multiple Skills coexisting does not increase Token consumption74% similarUnverified模型调用技能带有一段description，会进入Agent的上下文中，Agent可自行判断是否调用该技能73% similarUnverifiedAt the current stage, the Agent capabilities of the Active input method still require active user triggering and have not yet achieved fully automatic understanding and execution.72% similarUnverifiedSkill is a plugin mechanism for registering new callable tools with an Agent, allowing on-demand expansion of Agent capabilities without modifying the model itself.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/563121API
curl https://kongchang.com/api/v1/knowledge/claims/563121MCP
get_claim(id=563121)