← 返回论文库
MCP (Model Context Protocol)
Anthropic · 2024
L3.3 · LLM Agent Patterns & Frameworks
Open spec
#tool-use
#standard
CORE IDEA
agent-tool 通信的 HTTP for AI：server 暴露 resources/tools/prompts，client 标准化访问。
L-ANCHOR · 为什么在这一层重要
tool integration 标准
原文链接 ↗
源码 ↗
相关论文
ChemCrow: Augmenting Large-Language Models with Chemistry Tools
L1
2023
ReAct: Synergizing Reasoning and Acting in Language Models
L3.2
2022
Toolformer: Language Models Can Teach Themselves to Use Tools
L3.3
2023
OpenAI Function Calling
L3.3
2023