Unverified90% confidenceFactTime unknown
在Agent Scratchpad中，LLM生成的工具调用作为AI Message存储，工具返回的结果作为Tool Message存储，两者通过tool_call_id进行关联
1
Sources
90%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
LangChain Agent Executor构建原理：从零实现ReAct智能体
youtubeJames Briggs
Related Entities
Related Claims
UnverifiedWhen an LLM model decides to call a tool, it generates an assistant message containing a tool_use block; after the tool finishes executing, the result is appended as a new message in tool_result format73% similarUnverified在LangChain框架中，Tool调用结果通过ToolMessage对象追加到消息链68% similarUnverifiedHarness在转接过程中会预置系统提示词（system prompt），即便用户未写提示词，AI仍知道自己的身份65% similarUnverified动态工具机制允许智能体在运行过程中根据任务进展实时增删可用工具集65% similarUnverifiedAnthropic 的 Messages API 将 system 消息作为独立的顶层参数传递，而非放在 messages 数组中；工具调用使用 tool_use 和 tool_result content block 类型。64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/16130API
curl https://kongchang.com/api/v1/knowledge/claims/16130MCP
get_claim(id=16130)