Unverified50% confidenceOpinionExact time
MCP 方案的核心理念是让人预先编写可控代码暴露给智能体，由人决定智能体能执行什么，而非放任智能体直连系统自由生成 SQL
1
Sources
50%
Confidence
Long-term
Relevance
7/20/2026
First Seen
Sources
Related Claims
VerifiedMCP tools are essentially arbitrary code execution, indirect prompt injection is a real-world threat, and attack vectors can be hidden in resource content rather than in direct user input.77% similarUnverifiedMCP协议在工具调用安全性方面采用了人机协同（Human-in-the-Loop）的设计理念76% similarUnverifiedMCP 的在线发现机制是其区别于传统 function calling 方案的核心架构差异，使智能体在运行时动态了解可用能力而无需在训练阶段预先编码76% similarUnverifiedIn MCP, tools are functions that the model can execute, resources are read-only data with no side effects, and prompts are reusable templates that users actively select.76% similarUnverifiedMCP is responsible for connecting inanimate tools (databases, APIs) using synchronous calls; A2A is responsible for connecting intelligent agents using asynchronous collaboration.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/570003API
curl https://kongchang.com/api/v1/knowledge/claims/570003MCP
get_claim(id=570003)