Ollama Web Search MCP Service

A MCP (Model Context Protocol) service for performing web searches using Ollama.

Features

• Perform web searches using Ollama's search capabilities
• Format search results for easy consumption
• Official MCP Python SDK implementation
• Environment-based configuration
• Comprehensive error handling
• Stdio-based communication interface

Installation & Usage

Option 1: Direct execution with uvx (Recommended)

Install and run directly using uvx:

uvx ollama-websearch-mcp

This will automatically install the package and its dependencies, then start the MCP server.

Option 2: Local development

For local development, clone the repository and install dependencies:

git clone https://github.com/huangxinping/ollama-websearch-mcp.git
cd ollama-websearch-mcp
uv sync

Local usage commands

Run as MCP Server (for development):

python mcp_service.py

Build Package:

uv build

MCP Interface

Tools

1. search
   • Description: Perform a web search using Ollama
   • Parameters:
     • query (string): The search query
     • max_results (integer, optional): Maximum number of results to return (default: 3, range: 1-10)

Example Output

Search result format:

1. Example Title
   URL: https://example.com
   Content: Example content from the search result...

2. Another Title
   URL: https://another-example.com
   Content: Another example content...

Project Structure

ollama-websearch-mcp/
├── mcp_service.py             # MCP service main program
├── README.md                  # Project documentation
├── .env                       # Environment variables
├── .gitignore                 # Git ignore file
├── pyproject.toml             # Project configuration file
├── setup.py                   # Package setup file
└── uv.lock                    # Dependency lock file

Tech Stack

• Python 3.12+: Programming language
• MCP: Model Context Protocol framework
• Ollama: AI model API for web search
• python-dotenv: Environment variable management
• uv: Python package manager

Development Standards

• Use uv native commands for package management
• Follow Python PEP 8 coding standards
• Include type hints and docstrings
• Complete error handling and logging

Publishing to PyPI

To publish the package to PyPI, follow these steps:

1. Install development dependencies:

   uv sync --group dev
   

2. Build the package:

   uv build
   

3. Check the built package:

   uv run twine check dist/*
   

4. Upload to PyPI:

   uv run twine upload dist/*
   

Make sure you have your PyPI credentials configured before uploading.

AI IDE/CLI Configuration

Claude Code (CLI)

Add to your MCP configuration:

{
  "mcpServers": {
    "ollama-websearch": {
      "command": "uvx",
      "args": ["ollama-websearch-mcp"],
      "env": {
        "OLLAMA_API_KEY": "your_api_key_here"
      }
    }
  }
}

Cursor IDE

Add to your .cursorrules or MCP settings:

{
  "mcp": {
    "servers": {
      "ollama-websearch": {
        "command": "uvx",
        "args": ["ollama-websearch-mcp"],
        "env": {
           "OLLAMA_API_KEY": "your_api_key_here"
        }
      }
    }
  }
}

Windsurf IDE

Add to your Windsurf MCP configuration:

{
  "mcpServers": {
    "ollama-websearch": {
      "command": "uvx",
      "args": ["ollama-websearch-mcp"],
        "env": {
           "OLLAMA_API_KEY": "your_api_key_here"
        }
    }
  }
}

VS Code with Continue Extension

Add to your continue configuration:

{
  "mcp": {
    "servers": {
      "ollama-websearch": {
        "command": "uvx",
        "args": ["ollama-websearch-mcp"],
        "env": {
           "OLLAMA_API_KEY": "your_api_key_here"
        }
      }
    }
  }
}

Other MCP-Compatible Tools

For any MCP-compatible client, use:

# Command
uvx ollama-websearch-mcp

# Or with Python path
python -m mcp_service

License

MIT License