obsidian-rag-mcp

obsidian-rag-mcp

Exposes Obsidian notes as a semantic search and RAG knowledge base over MCP, enabling AI assistants to index, retrieve, and analyze personal notes via natural language.

Category
Visit Server

README

obsidian-rag-mcp

一个基于 MCP (Model Context Protocol) 的 RAG 服务器:把 Obsidian 笔记库变成 AI(goose)可检索的知识库。

用户在对话中输入 /obsidian-rag <内容>,goose 会先从你的 Obsidian 笔记中语义检索相关片段,再结合这些片段分析你的内容——让你的笔记成为 AI 的「第二大脑」。

✨ 功能

  • 📁 读取 Obsidian vault:扫描 *.md 笔记,自动忽略 .obsidian.trash.git 等隐藏目录
  • 🧠 可配置 Embedding 模型:支持 OpenAI 兼容 API 与 Ollama 本地模型(也内置 fake 模式用于零依赖测试)
  • 🔍 语义检索:纯 Python 余弦相似度,无需重型向量数据库
  • 🧩 MCP 标准协议:可接入 goose / Claude / 任何 MCP 客户端
  • 🔧 7 个工具:索引、搜索、RAG 检索、列笔记、读笔记、查配置、查索引状态

🛠 工具一览

工具 说明
obsidian_index(force) 扫描 vault 并构建/重建 embedding 索引
obsidian_search(query, top_k) 语义搜索笔记片段
obsidian_rag(question, top_k) 检索与问题最相关的笔记上下文(供分析)
obsidian_list_notes(keyword) 列出 vault 中的笔记
obsidian_read_note(path) 读取单篇笔记全文(防路径穿越)
obsidian_get_config() 查看当前配置(不含 API Key)
obsidian_index_status() 检查索引是否存在且模型匹配

🚀 快速开始

1. 克隆并安装

git clone https://github.com/<your-org>/obsidian-rag-mcp.git
cd obsidian-rag-mcp
uv sync

2. 配置环境变量

在 goose 中添加扩展时配置(或在终端导出):

变量 必填 说明 默认值
OBSIDIAN_VAULT_PATH Obsidian vault 的绝对路径
EMBEDDING_BASE_URL Embedding API 地址(Ollama 用 http://localhost:11434 无地址且无 key 时为离线测试模式
EMBEDDING_MODEL Embedding 模型名 text-embedding-3-small(OpenAI 兼容)
EMBEDDING_API_KEY OpenAI 兼容时必填 API Key(Ollama 本地无需)
OBSIDIAN_INDEX_PATH 索引文件保存位置 ~/.obsidian-rag/index.json
OBSIDIAN_CHUNK_SIZE 分块字符数 1500
OBSIDIAN_MAX_NOTES 最多索引的笔记数 1000

后端自动识别:只需配置地址 + 模型 + key,无需指定 provider。 地址含 Ollama 默认端口 11434 或以 /api 结尾 → 自动按 Ollama 调用; 其他地址 → 自动按 OpenAI 兼容 POST {base}/embeddings 调用。

3. 在 goose 中注册扩展

config.yaml(Windows: %APPDATA%\Block\goose\config\config.yaml)的 extensions 下添加:

extensions:
  obsidian-rag:
    type: stdio
    name: obsidian-rag
    enabled: true
    cmd: uv
    args:
      - run
      - --directory
      - "D:/path/to/obsidian-rag-mcp"
      - obsidian-rag-mcp
    envs:
      OBSIDIAN_VAULT_PATH: "D:/path/to/your/vault"
      EMBEDDING_BASE_URL: "https://api.openai.com/v1"
      EMBEDDING_MODEL: "text-embedding-3-small"
      EMBEDDING_API_KEY: "<你的 key>"
    timeout: 300

也可用 goose configureAdd ExtensionCommand-Line Extension 交互式添加。

4. 配置 /obsidian-rag 命令

config.yaml 中添加:

slash_commands:
  - command: "obsidian-rag"
    recipe_path: "D:/path/to/obsidian-rag-mcp/recipes/obsidian-rag.yaml"

重启会话后,在对话中输入 /obsidian-rag 帮我分析一下我对新项目的想法 即可。

5. 直接使用(无需命令)

goose 会自动决定何时调用工具,你也可以直接要求:

「用我的 Obsidian 笔记分析一下这个方案的可行性」「检索我笔记里关于网站改版的内容」

🔬 本地测试

uv run --directory . pytest

测试使用内置 fake embedding(确定性哈希向量),无需任何 API Key 和网络即可端到端验证整个 RAG 流程(扫描 → 分块 → 索引 → 检索 → MCP 调用)。

💡 Embedding 配置示例

OpenAI 兼容 API(如自建代理 / 中转)

export EMBEDDING_BASE_URL=https://your-gateway/v1
export EMBEDDING_MODEL=text-embedding-3-small
export EMBEDDING_API_KEY=sk-xxx

Ollama 本地模型

ollama pull nomic-embed-text
export EMBEDDING_BASE_URL=http://localhost:11434
export EMBEDDING_MODEL=nomic-embed-text

离线测试模式(无需网络 / 无需 key)

# 什么都不配置即可(检测不到地址和 key 时自动进入该模式)

📂 项目结构

obsidian-rag-mcp/
├── obsidian_rag/
│   ├── server.py       # MCP 服务器与工具定义
│   ├── config.py       # 环境变量配置
│   ├── embeddings.py   # OpenAI/Ollama/fake embedding 客户端
│   ├── vault.py        # vault 扫描与 markdown 分块
│   └── store.py        # 向量存储与余弦相似度检索
├── recipes/
│   └── obsidian-rag.yaml   # /obsidian-rag slash command recipe
├── examples/sample-vault/  # 示例笔记库
└── tests/

📄 License

MIT

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured