How do I install minirag-mcp?

Pick a client above and copy the generated config. Cursor and VS Code also have a deep link when the server can run locally or remotely.

Is minirag-mcp official?

No. Registry-listed is not the same as vendor-official. Read the methodology for the difference.

How is the trust score calculated?

This listing scores 50 (Fair). The formula is published on the methodology page.

MCPpedia

minirag-mcp MCP Server

Name: minirag-mcp MCP Server
Author: sfrangulov

Local-first RAG MCP server: hybrid search over a folder of your own documents

Tools Versions Alternatives

Registrysfrangulov

50Fair

Install

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "sfrangulov-minirag-mcp": {
      "command": "uvx",
      "args": [
        "minirag-mcp"
      ],
      "env": {
        "BASE_DIR": "",
        "BASE_DIRS": "",
        "DB_PATH": "",
        "CACHE_DIR": "",
        "MODEL_NAME": "sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
        "MAX_FILE_SIZE": "104857600",
        "CHUNK_TOKEN_BUDGET": "110",
        "RAG_HYBRID_WEIGHT": "0.6",
        "RAG_GROUPING": "",
        "RAG_MAX_DISTANCE": "",
        "RAG_MAX_FILES": "",
        "RAG_INSTRUCTIONS_APPEND": "",
        "ALLOW_PRIVATE_URLS": "false"
      }
    }
  }
}

Install guide

Related servers

SuperLocalMemory

78Good

Local-first memory for AI agents: recall with sources, answer checks, documents, mesh, governance.

Registryqualixar249 stars1.3K/wk

Codebase Memory

77Good

Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.

minirag-mcp MCP Server | MCPpedia

Environment

Name	Required	Notes
BASE_DIR	No	One document root; also the security boundary for file access. Defaults to the process working directory. Ignored when BASE_DIRS is set.
BASE_DIRS	No	JSON array of document roots, e.g. ["/docs/a", "/docs/b"]. Takes precedence over BASE_DIR. An invalid value is a hard configuration error.
DB_PATH	No	LanceDB index directory. Defaults to <first root>/.minirag/lancedb, so each corpus gets its own index.
CACHE_DIR	No	Embedding model cache. Defaults to the platform user cache dir, so the ~220 MB model is downloaded once and shared.
MODEL_NAME	No	fastembed model id. Changing it makes existing vectors incompatible with new queries; pair with a new DB_PATH or a full re-ingest.
MAX_FILE_SIZE	No	Per-file size limit in bytes, enforced before parsing.
CHUNK_TOKEN_BUDGET	No	Retrieval-unit size in the embedding model's own tokens. Range 16-128; the ceiling is the model's trained sequence length.
RAG_HYBRID_WEIGHT	No	Keyword weight in the weighted RRF fusion, range 0.0-1.0. 0 is vector-only; higher values raise the BM25 contribution.
RAG_GROUPING	No	Result grouping filter. 'similar' keeps only the closest group; 'related' also keeps the next one. Unset means no grouping filter.
RAG_MAX_DISTANCE	No	Drop results whose vector distance exceeds this value. Lower is stricter. Unset means no distance filter.
RAG_MAX_FILES	No	Keep chunks from at most this many best-scoring files. Unset means no per-file filter.
RAG_INSTRUCTIONS_APPEND	No	Extra paragraph appended to the instructions the server hands the client at connect time, for corpus-specific guidance.
ALLOW_PRIVATE_URLS	No	Let ingest_url fetch hosts resolving to loopback, link-local, private, reserved or unspecified addresses. Off by default.

minirag-mcp MCP Server

Install

Related servers

SuperLocalMemory

Codebase Memory

Environment

Tools

About

FAQ

Reviews

Firecrawl MCP Server

Zotero MCP

haiku.rag

openInvest