About Fess MCP Server
Fess MCP Server connects MCP clients and agents to the Fess full-text search engine. It lets agents retrieve information through Fess when registered in a compatible client.
Key features
- 01Connects to Fess search
- 02Enables agent information retrieval
- 03Configurable Fess server URL
- 04Docker or uv setup
Use cases
- 01Retrieve information through Fess from an MCP client
- 02Provide agents access to Fess search
MCP support
It exposes Fess search to MCP clients, allowing connected agents to obtain information through the search engine.
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 READMEFess MCP Server
Fess MCP Server is a middleware server that integrates with the Fess search engine. By registering it in the settings of MCP clients such as Claude for Desktop, it enables agents to obtain information using Fess. Fess is an open-source full-text search server provided under the Apache License. Commercial support is also available. For more details, please refer to the official website. https://fess.codelibs.org/ja/
Setup
Fess Setup
For the setup procedure of the Fess server itself, please refer to the official Fess documentation. https://qiita.com/g_fukurowl/items/a00dbbad737d3e775108
The following article may also be helpful: https://qiita.com/g_fukurowl/items/a00dbbad737d3e775108
Fess MCP Server Setup
Using Docker
Start the server with the following command:
docker-compose up -d
Without Docker
Here we show the setup procedure using uv. uv is a fast Python package manager. While not required, we recommend using it.
# Install uv
irm https://astral.sh/uv/install.ps1 | iex
# Activate virtual environment
.\.venv\Scripts\activate.bat
# Install dependencies
uv pip install -e .
# Start MCP server
uv run .\fess_mcp_server.py
Configuration
Fess Server Connection Settings
The URL of the Fess server that Fess MCP Server accesses is set as the environment variable Fess_API_BASE. Please modify this variable according to your environment before starting Fess MCP Server. If running as a Docker container, edit docker-compose.yaml and change the relevant settings.
Claude for Desktop Connection Settings
Fess MCP Server starts on port 8000 by default. If Fess MCP Server is running on localhost on port 8000, edit claude_desktop_config.json as follows:
{
"mcpServers": {
"fess-search-sse": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"http://localhost:8000/sse"
]
}
}
}
Testing
Without Docker
Here we show the testing procedure using uv.
# Install development dependencies:
uv pip install -e ".[test]"
# Run unit tests:
uv run pytest -v tests/unit/ -s
# Run integration tests (requires communication with Fess server):
uv run pytest -v tests/integration/ -s
# Generate coverage report:
uv run pytest --cov=fess_mcp_server --cov-report=html
Running Tests in Docker Container
# Build test Docker image:
docker build -t fess-mcp-server-test -f Dockerfile.test .
# Run unit tests:
docker run --rm fess-mcp-server-test pytest -v tests/unit/ -s
# Run integration tests (requires communication with Fess server):
docker run --rm --network host fess-mcp-server-test pytest -v tests/integration/ -s
# Generate coverage report:
docker run --rm -v ${PWD}/coverage:/app/coverage fess-mcp-server-test pytest --cov=fess_mcp_server --cov-report=html
Test Execution Options
-v: Show verbose output-s: Show standard output--cov: Generate coverage report--cov-report=html: Generate HTML coverage report
License
MIT License
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.