How do I use docs2db-mcp-server?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@docs2db-mcp-server search for steps to configure SELinux" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

docs2db-mcp-server

Official

by rhel-lightspeed

Overview Schema Source Related Servers Discussions Admin

Python

Hybrid

docs2db-mcp-server

MCP server for docs2db-api RAG queries. Provides semantic and hybrid search tools for RHEL documentation. Compatible with llama-stack, Goose, Claude Desktop, and other MCP clients.

Features

Hybrid Search: Combines vector similarity and BM25 keyword search with Reciprocal Rank Fusion
Cross-Encoder Reranking: Improves result quality using ms-marco-MiniLM-L-6-v2
MCP Protocol: Standard Model Context Protocol via FastMCP
SSE Transport: Server-Sent Events for real-time streaming
Configurable: Environment variables for all settings
Production Ready: Containerized, non-root user, health checks

Related MCP server: hybrid-rag-mcp

Installation

Prerequisites

Install uv:

# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Or with brew
brew install uv

From PyPI

uv add docs2db-mcp

From Source

git clone https://github.com/rhel-lightspeed/docs2db-mcp-server.git
cd docs2db-mcp-server
uv sync
uv pip install -e .

Quick Start

Prerequisites

PostgreSQL with pgvector containing docs2db RAG database
docs2db to build the RAG database (see docs2db)

Running the Server

# Start the MCP server (uses localhost defaults)
uv run python -m docs2db_mcp

Default configuration connects to postgresql://postgres:postgres@localhost:5432/ragdb

Server will start on http://localhost:8002/sse

Using with llama-stack

Add to your run.yaml:

tool_runtime:
  - provider_id: model-context-protocol
    provider_type: remote::model-context-protocol

registered_resources:
  tool_groups:
    - toolgroup_id: mcp::docs2db-rag
      provider_id: model-context-protocol
      mcp_endpoint:
        uri: http://localhost:8002/sse

Using with Goose

Add to ~/.config/goose/config.yaml:

extensions:
  docs2db-rag:
    enabled: true
    type: stdio
    name: docs2db-rag
    cmd: uv
    args:
    - run
    - python
    - -m
    - docs2db_mcp
    envs:
      DOCS2DB_MCP_TRANSPORT: stdio

Using with Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "docs2db-rag": {
      "command": "uv",
      "args": ["run", "python", "-m", "docs2db_mcp"]
    }
  }
}

Configuration

All configuration via environment variables with DOCS2DB_MCP_ prefix.

Note: For stdio transport, DOCS2DB_LOG_LEVEL is automatically set to CRITICAL to minimize startup logs. Set manually if needed.

MCP Server Settings

Variable	Default	Description
`DOCS2DB_MCP_TRANSPORT`	`sse`	Transport type (sse/stdio)
`DOCS2DB_MCP_HOST`	`0.0.0.0`	Bind address
`DOCS2DB_MCP_PORT`	`8002`	Port number
`DOCS2DB_MCP_LOG_LEVEL`	`INFO`	Logging level
`DOCS2DB_MCP_SHOW_BANNER`	`true`	Show FastMCP banner (auto-suppressed for stdio)

Database Settings

Variable	Default	Description
`DOCS2DB_MCP_DB_HOST`	`localhost`	PostgreSQL host
`DOCS2DB_MCP_DB_PORT`	`5432`	PostgreSQL port
`DOCS2DB_MCP_DB_DATABASE`	`ragdb`	Database name
`DOCS2DB_MCP_DB_USER`	`postgres`	Database user
`DOCS2DB_MCP_DB_PASSWORD`	`postgres`	Database password

RAG Settings

Variable	Default	Description
`DOCS2DB_MCP_RAG_SIMILARITY_THRESHOLD`	`0.7`	Minimum similarity score
`DOCS2DB_MCP_RAG_MAX_CHUNKS`	`5`	Maximum chunks to return
`DOCS2DB_MCP_RAG_ENABLE_RERANKING`	`true`	Enable cross-encoder reranking

Tools

search_documents

Search RHEL documentation using hybrid search (vector + BM25).

Parameters:

query (string, required): Search query
max_chunks (integer, optional): Maximum chunks to return (default: 5)
similarity_threshold (float, optional): Minimum similarity score (default: 0.7)
enable_reranking (boolean, optional): Enable cross-encoder reranking (default: true)

Returns:

{
  "chunks": [
    {
      "text": "Chunk content...",
      "contextual_text": "LLM-generated context...",
      "similarity": 0.85,
      "source": "path/to/source/file.html",
      "metadata": {...}
    }
  ],
  "query_used": "original query"
}

Docker/Podman

Build Image

podman build -t docs2db-mcp-server .

Run Container

podman run -d \
  --name docs2db-mcp \
  -p 8002:8002 \
  -e DOCS2DB_MCP_DB_HOST=postgres \
  -e DOCS2DB_MCP_DB_PASSWORD=mysecret \
  docs2db-mcp-server

Docker Compose / Podman Compose

services:
  postgres:
    image: pgvector/pgvector:pg17
    environment:
      POSTGRES_PASSWORD: postgres
    volumes:
      - ragdb:/var/lib/postgresql/data
      - ./ragdb_dump.sql:/docker-entrypoint-initdb.d/init.sql

  docs2db-mcp:
    image: docs2db-mcp-server
    ports:
      - "8002:8002"
    environment:
      DOCS2DB_MCP_DB_HOST: postgres
      DOCS2DB_MCP_DB_PASSWORD: postgres
    depends_on:
      - postgres

Development

# Clone repo
git clone https://github.com/rhel-lightspeed/docs2db-mcp-server.git
cd docs2db-mcp-server

# Install dependencies
uv sync
uv pip install -e ".[dev]"

# Run tests
uv run pytest

# Run linters
uv run ruff check .
uv run mypy src/

Architecture

┌─────────────────────────────────────┐
│ MCP Client                          │
│ (llama-stack, Goose, Claude, etc.)  │
└──────────────┬──────────────────────┘
               │ MCP Protocol (SSE/stdio)
               ↓
┌─────────────────────────────────────┐
│ docs2db-mcp-server                  │
│                                     │
│ FastMCP Server                      │
│   └─ search_documents (tool)        │
│         ↓                           │
│   UniversalRAGEngine                │
│   (from docs2db-api)                │
└──────────────┬──────────────────────┘
               │
               ↓
┌─────────────────────────────────────┐
│ PostgreSQL + pgvector               │
│ (docs2db RAG database)              │
└─────────────────────────────────────┘

docs2db - Build RAG databases from documents
docs2db-api - Query API for docs2db databases
linux-mcp-server - MCP server for Linux diagnostics

License

Apache License 2.0 - See LICENSE for details.

Contributing

Contributions welcome! Please open an issue or pull request.

Support

Issues: https://github.com/rhel-lightspeed/docs2db-mcp-server/issues
Discussions: https://github.com/rhel-lightspeed/docs2db-mcp-server/discussions

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivitySlowing

ResponsivenessUnresponsive

Resources

GitHub Repository

Need Help?

Related Servers

Related MCP Connectors

mcp A
ai.llmse
Public MCP server for the LLM Search Engine
Search RAG Systems Developer Tools
102
forkast-mcp-docs A
io.github.ForkastPredict
MCP server for querying Forkast documentation
Documentation Access Search Developer Tools
42
docbaseOAuth
io.github.krayinc
DocBase MCP server for AI agents
Documentation Access Search Developer Tools
2
DocsMint A
io.github.HiAi-gg
Cloud or self-hosted knowledge for AI agents: hybrid search, reranking, GraphRAG, scoped MCP tools.
Documentation Access Knowledge & Memory RAG Systems
21171

Related MCP Servers

LLMDoc
RAG Systems Search Documentation Access
bigbag
A
license
A
quality
D
maintenance
MCP server for semantic search across llms.txt documentation sources, with hybrid two-stage retrieval and automatic background refresh.
Updated 8 months ago (2026-01-23 06:59 UTC)
5
MIT
hybrid-rag-mcp
RAG Systems Search
udarshmarthala
A
license
A
quality
C
maintenance
An MCP server that provides a search_docs tool with hybrid retrieval (BM25 + dense vectors) and cross-encoder reranking, backed by evaluation, prompt-injection guardrails, and OpenTelemetry tracing.
Updated a month ago (2026-08-11 02:51 UTC)
1
MIT
rag-mcp-server
RAG Systems Search Vector Databases
weitingyen168
A
license
Not graded
quality
C
maintenance
RAG document search MCP server that allows AI assistants to search a local document set and retrieve grounded passages via keyword (SQLite FTS5) or semantic (Chroma) backends.
Updated 24 days ago (2026-08-27 08:27 UTC)
MIT
mcp-rag
RAG Systems Vector Databases Search
nguyenkhanhtoan-ai
F
license
Not graded
quality
B
maintenance
MCP server that enables semantic search over PDF documents using RAG with ChromaDB and OpenAI embeddings.
Updated a month ago (2026-08-14 09:06 UTC)
-

docs2db-mcp-server

Features

Installation

Prerequisites

From PyPI

From Source

Quick Start

Prerequisites

Running the Server

Using with llama-stack

Using with Goose

Using with Claude Desktop

Configuration

MCP Server Settings

Database Settings

RAG Settings

Tools

search_documents

Docker/Podman

Build Image

Run Container

Docker Compose / Podman Compose

Development

Architecture

Related Projects

License

Contributing

Support

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

LLMDoc

hybrid-rag-mcp

rag-mcp-server

mcp-rag