Ollama MCP Bridge WebUI

🚀 Ollama-MCP Bridge WebUI

A bridge implemented in TypeScript that connects local LLMs (via Ollama) to Model Context Protocol (MCP) servers and provides a web interface. This bridging tool enables open-source models to have the same capabilities as Claude, thus realizing a powerful local AI assistant that runs entirely on your own hardware.

✨ Features

• Multi-MCP Integration: Connect to multiple MCP servers simultaneously.
• Tool Recognition: Automatically identify the tools to be used based on the query.
• Web Interface: A clean user interface with collapsible tool descriptions.
• Comprehensive Toolset: File system, web search, and inference capabilities.

📦 Installation

Automatic Installation

The easiest way to set up the bridging tool is to use the included installation script:

./install.bat

The script will perform the following actions:

1. Check and install Node.js if needed.
2. Check and install Ollama if needed.
3. Install all dependencies.
4. Create a workspace directory (../workspace).
5. Set up the initial configuration.
6. Build the TypeScript project.
7. Download the Qwen model to Ollama.

After running the script, you only need to do the following:

1. Add the API key to the .env file (the $VARIABLE_NAME variables in the configuration file will be replaced with actual values).

Manual Setup

If you prefer manual setup:

1. Install Ollama from ollama.com/download.
2. Pull the Qwen model: ollama pull qwen2.5-coder:7b-instruct-q4_K_M.
3. Install dependencies: npm install.
4. Create a workspace directory: mkdir ../workspace.
5. Configure the API key in the .env file.
6. Build the project: npm run build.

📚 Documentation

The bridging tool is configured through two main files:

1. bridge_config.json

This file defines MCP servers, LLM settings, and system prompts. Use the $VARIABLE_NAME syntax to reference environment variables.

Example:

{
  "mcpServers": {
    "filesystem": {
      "command": "node",
      "args": [
        "To/Your/Directory/Ollama-MCP-Bridge-WebUI/node_modules/@modelcontextprotocol/server-filesystem/dist/index.js",
        "To/Your/Directory/Ollama-MCP-Bridge-WebUI/../workspace"
      ],
      "allowedDirectory": "To/Your/Directory/Ollama-MCP-Bridge-WebUI/../workspace"
    },
    "brave-search": {
      "command": "node",
      "args": [
        "To/Your/Directory/Ollama-MCP-Bridge-WebUI/node_modules/@modelcontextprotocol/server-brave-search/dist/index.js"
      ],
      "env": {
        "BRAVE_API_KEY": "$BRAVE_API_KEY"
      }
    },
    "sequential-thinking": {
      "command": "node",
      "args": [
        "To/Your/Directory/Ollama-MCP-Bridge-WebUI/node_modules/@modelcontextprotocol/server-sequential-thinking/dist/index.js"
      ]
    }
  },
  "llm": {
    "model": "qwen2.5-coder:7b-instruct-q4_K_M",
    "baseUrl": "http://localhost:11434",
    "apiKey": "ollama",
    "temperature": 0.7,
    "maxTokens": 8000
  },
  "systemPrompt": "You are a helpful assistant."
}

2. .env

Store environment variables, such as API keys:

BRAVE_API_KEY=your_brave_api_key_here

💻 Usage Examples

Basic Usage

Run the bridging tool:

./start.bat

Then access the local web page in your web browser.

📄 License

[Add license information here, e.g., MIT or GPL]

🤝 Contribution

Contributions are welcome! Please check CONTRIBUTING.md for more information.

📞 Contact

• GitHub: https://github.com/yourusername/ollama-mcp-bridge
• Email: contact@yourdomain.com
• Discussion Group: Join our Discord or other chat groups