velocirag — Claude Code MCP server: what it does, how to install
A local RAG server that lets agents search indexed documents through MCP. It combines semantic, keyword, graph, and metadata retrieval.
claude mcp add velocirag -- uvx velociragWhat velocirag does
velocirag builds a document index and combines four retrieval paths: vector similarity, BM25 keyword search, knowledge-graph traversal, and metadata filters. Rank fusion and a cross-encoder then reorder the results. Its MCP interface gives an agent tools for search, indexing, single-document insertion, diagnostics, and source listing.
The stack uses ONNX Runtime, FAISS, SQLite FTS5, and local MiniLM and TinyBERT models. Files can be indexed incrementally, while provenance tracking updates affected graph links. A Unix-socket daemon keeps the search engine warm between CLI requests. It runs on CPU and requires 8GB RAM; GPU hardware and API keys are not needed.
What it helps with
- index a directory of documents
- find exact terms and related concepts
- filter results by tags and status
- add one document to the knowledge base
- check MCP server health
Example prompts for the agent
- Index
./docsand find material about schema migrations. - Add
README.mdto the knowledge base, then check the server health. - Search for documents tagged
pythonwith statusactivethat relate to error handling.
How to install velocirag
Platform CLI
claude mcp add velocirag -- uvx velociraguvx runs the command named after the package; if the README names another one, use uvx --from velocirag <command>
Platform CLI
codex mcp add velocirag -- uvx velociraguvx runs the command named after the package; if the README names another one, use uvx --from velocirag <command>
Config file
{
"mcpServers": {
"velocirag": {
"command": "uvx",
"args": [
"velocirag"
]
}
}
}uvx runs the command named after the package; if the README names another one, use uvx --from velocirag <command>
Config file
{
"servers": {
"velocirag": {
"type": "stdio",
"command": "uvx",
"args": [
"velocirag"
]
}
}
}uvx runs the command named after the package; if the README names another one, use uvx --from velocirag <command>
Platform CLI
gemini mcp add velocirag uvx velociraguvx runs the command named after the package; if the README names another one, use uvx --from velocirag <command>
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about velocirag
How do I install velocirag in Claude Code?
Run claude mcp add velocirag -- uvx velocirag.
What does velocirag do?
It indexes documents and searches them through vector, keyword, graph, and metadata retrieval. Agents can call tools for search, indexing, document insertion, diagnostics, and source listing.
Does velocirag require a GPU or API key?
No. Search uses CPU-based ONNX Runtime, and the server requires 8GB RAM.
