About Wandering RAG
Wandering RAG is a CLI for personal RAG that indexes content from sources such as Notion, Obsidian, and Apple Notes in Qdrant. It also runs an MCP server for accessing that content, with Markdown indexing documented and Notion indexing marked as work in progress.
Key features
- 01Index Markdown folders
- 02Support Obsidian vaults
- 03Notion indexing marked as work in progress
- 04Store indexed data in Qdrant
- 05Run an MCP server
Use cases
- 01Search personal Markdown notes with RAG
- 02Access indexed notes from Claude Desktop
MCP support
Its CLI can run an MCP server that provides access to personal RAG data stored in Qdrant.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEWandering RAG
A CLI tool for personal RAG that retrieves data from Notion, Obsidian, Apple Notes, etc., stored in Qdrant and exposed as an MCP server.
(So that Claude Desktop successfully answers my question "When did I adopt my cat and when did I change her cat litter most recently? ")
Installation
Using uv (recommended):
uv pip install -e .
Run a Qdrant server if you don't have one:
cd qdrant-docker
docker-compose up -d
Copy .env.example as .env and specify the Markdown folders (or Obsidian vaults)
Usage
The CLI provides several subcommands:
Markdown commands
./wandering-rag md index
Notion commands (WIP)
./wandering-rag notion index
MCP commands
./wandering-rag mcp run-server
Configure for Claude Desktop
{
"mcpServers": {
"wandering-rag": {
"command": "<your git checkout path>/wandering-rag/wandering-rag",
"args": ["mcp", "run-server"],
"env": {
"PATH": "<your home folder>/.local/bin:/usr/bin:$PATH"
}
}
}
}
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.