GopherDigest

An MCP server that reads conversation transcripts from disk, sends them to a local Ollama instance for fact extraction, and merges the results into a GopherCache context.json file.

Prerequisites

• Python 3.10+
• Ollama running locally on http://localhost:11434
• The llama3.2 model pulled in Ollama:

ollama pull llama3.2

Ollama must be running before any tool is called. Start it with ollama serve if it is not already running as a background service.

Installation

# Clone / enter the repo
cd gopherdigest

# Create and activate a virtual environment
python -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate

# Install the package
pip install -e .

Copy .env.example to .env and adjust if needed:

cp .env.example .env

Tools

Configuration

Claude Desktop configuration

Add the following to your Claude Desktop claude_desktop_config.json (usually at ~/Library/Application Support/Claude/claude_desktop_config.json on macOS or %APPDATA%\Claude\claude_desktop_config.json on Windows):

{
  "mcpServers": {
    "gopherdigest": {
      "command": "python",
      "args": ["-m", "gopherdigest.server"],
      "cwd": "/absolute/path/to/gopherdigest",
      "env": {
        "GOPHERDIGEST_DATA_DIR": "./data"
      }
    }
  }
}

Replace /absolute/path/to/gopherdigest with the actual path where you cloned this repo.

If you installed into a virtual environment, point command at the venv interpreter instead:

{
  "mcpServers": {
    "gopherdigest": {
      "command": "/absolute/path/to/gopherdigest/.venv/bin/python",
      "args": ["-m", "gopherdigest.server"],
      "cwd": "/absolute/path/to/gopherdigest"
    }
  }
}

Running the server manually (for testing)

python -m gopherdigest.server

Or via the installed script:

gopherdigest

License

MIT — see LICENSE.