MCP -readability-python

This MCP server implements a Python-based Readability parser that extracts the main article content from web pages and converts it into clean, Markdown-formatted text optimized for language models. It runs as a FastMCP server and exposes a tool named extract_content under the MCP protocol. To use it, start the MCP server with the configured command, then issue a parse-style request to the /tools/extract_content endpoint with the target URL. The server handles noise removal (ads, navigation, footers) and formats the result as Markdown suitable for downstream LLM processing.

Prerequisites:

• Python 3.8+ and a working virtual environment manager (venv).
• Git (to clone the repository) if you are starting from source.
• Internet access to install dependencies from PyPI.

Step-by-step installation:

1. Clone the repository:

git clone https://github.com/jmh108/MCP-server-readability-python.git
cd MCP-server-readability-python

2. (Optional) Create and activate a virtual environment:

python -m venv venv
# macOS/Linux
source venv/bin/activate
# Windows
venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Run the server using FastMCP (as configured in MCP config):

fastmcp run server.py

Notes:

• Ensure port 8000 is available or configure FastMCP as needed in your environment.
• If you need to customize environment variables, modify the mcp configuration (env section) accordingly.

Tips and common considerations:

• The extract_content tool expects a JSON payload with the field url, e.g., {"url": "https://example.com/article"}.
• The server uses readability-lxml for content extraction and html2text for Markdown conversion; ensure network access to fetch pages and to install dependencies.
• If you update dependencies, re-run pip install -r requirements.txt to keep the environment in sync.
• You can adjust MCP settings to tune timeouts or add additional tools under the same mcpServers block.
• Common issues: port conflicts, missing dependencies, or firewall rules blocking outbound requests. Check server logs for startup errors and verify the server is listening on the expected port.
• Environment variables can be added in the env section of the mcpServers configuration to customize behavior (e.g., proxy settings, timeouts).