Model Context Protocol
Core Facts
When connecting to multiple MCP servers simultaneously, tool descriptions alone can consume thousands or even tens of thousands of tokens.
MCP defines three core primitives: Tools (tool invocation), Resources (resource reading), and Prompts (prompt templates).
MCP采用客户端-服务器架构，基于标准化的JSON-RPC通信协议
Timeline (last 90 days)
The MCP ecosystem is still evolving rapidly, and selection should be based on actual project judgment rather than copying a one-size-fits-all approach.
The model relies entirely on a tool's description field to judge when to call it, and vague descriptions lead to erroneous or missed calls.
It is suggested that initially MCP tools be made fine-grained with single responsibilities, paired with a permission policy that separates read and write operations, and then iteratively converged based on actual call logs.
For teams connecting AI agents into production environments, MCP is not optional but infrastructure.
The recommended path for connecting AI to Home Assistant is: enable MCP, start read-only, make backups, learn approaches from cloud models, and then implement with local models.
The community version of MCP has delete permissions by default, and the author recommends setting it to read-only initially, backing up first, and then enabling write permissions.
The official Home Assistant integration's MCP provides about seven tools, while the community version provides more than seventy tools.
For enterprises evaluating property management software, it is recommended to include 'whether it supports MCP and open APIs' in the procurement checklist.
PaperPilot is a pure Rust PDF engine that runs entirely locally, executes in the browser via WASM, and exposes its capabilities to AI tools through the MCP protocol.
Anthropic had previously open-sourced MCP and pushed it to become a de facto standard, and srt's open posture is consistent with that.
40 more timeline events
All Facts (20)
When connecting to multiple MCP servers simultaneously, tool descriptions alone can consume thousands or even tens of thousands of tokens.
90%VerifiedMCP defines three core primitives: Tools (tool invocation), Resources (resource reading), and Prompts (prompt templates).
90%VerifiedMCP采用客户端-服务器架构，基于标准化的JSON-RPC通信协议
90%VerifiedAnthropic launched MCP (Model Context Protocol) at the end of 2024, an open standard designed to standardize the way AI models interact with external tools and data sources.
90%VerifiedMCP standardizes the integration complexity between AI applications and external tools from M×N to M+N.
90%VerifiedIn MCP, Tools are invoked at the model's discretion, Prompts are triggered by the user, and Resources are controlled by the application.
80%VerifiedMCP协议支持两种传输方式：stdio（标准输入输出）和HTTP SSE（Server-Sent Events）
80%VerifiedMCP采用客户端-服务器架构，AI编程工具作为MCP客户端，外部服务作为MCP服务器
80%VerifiedMCP基于JSON-RPC 2.0协议构建，支持请求-响应和通知两种消息模式
80%VerifiedMCP（Model Context Protocol）是Anthropic提出的开放协议，旨在标准化大模型与外部工具、数据源之间的交互方式
80%VerifiedMCP（Model Context Protocol）是Anthropic于2024年底推出的开放标准，采用客户端-服务器架构和JSON-RPC协议通信
80%VerifiedMCP的架构由三层组成：Host（宿主应用）、Client（客户端）和Server（服务端）
80%VerifiedMCP（模型上下文协议）是由Anthropic于2024年底推出的开放标准协议，旨在为AI模型与外部工具、数据源之间建立统一的通信接口
80%VerifiedThe Model Context Protocol (MCP) is an open protocol officially open-sourced and released by Anthropic in November 2024.
80%VerifiedMCP (Model Context Protocol) is an open standard protocol released by Anthropic in late 2024 that defines a unified communication specification between AI models and external tools and data sources.
80%VerifiedMCP（Model Context Protocol）是由 Anthropic 于2024年底提出并开源的标准化协议，旨在让 AI 模型与外部工具、数据源之间的交互有统一的接口规范
80%VerifiedBlender MCP is a plugin implemented based on the MCP protocol that exposes Blender's Python API to language models, enabling the model to create objects, set keyframes, and adjust cameras.
75%VerifiedMCP addresses the connectivity problem (how a model reaches tools), while Skill addresses the usage problem (when and with what strategy the model combines tools).
75%VerifiedOne of the security risks of MCP is prompt injection attacks, where malicious data sources may embed instructions in returned content to induce the model to perform unauthorized operations.
70%VerifiedThe MCP authorization specification emphasizes token audience validation, requiring MCP servers to verify that a token was indeed issued to themselves in order to prevent confused deputy attacks.
70%