Internetsearch

You run an MCP server that uses BoCha AI to perform internet searches and return results to clients. It’s useful for building search-backed assistants, bots, or data tools that rely on real-time web information via a dedicated MCP endpoint.

How to use

You access this MCP server from your MCP client by configuring a local stdio-based MCP entry that launches the server process. The server talks to BoCha AI’s search API and returns structured results that your client can present or further process. Use this setup whenever you want a reusable, self-contained Internet search source inside your MCP ecosystem.

What you can do with it

• Trigger web searches and retrieve results through the MCP interface
• Integrate search results into your workflows, bots, or automated agents
• Centralize internet search capabilities behind a single MCP endpoint for consistent usage across clients.

How to install

# Prerequisites
# Ensure you have Python installed
python --version

# Prerequisite package for MCP runtime
pip install uv

# Clone the project
git clone https://github.com/mingdedi/Internetsearch-mcp-server.git

# Change into the project directory
cd Internetsearch-mcp-server

# Recreate the Python virtual environment for the MCP runtime
uv venv

# Activate the virtual environment
# On Windows
./.venv/Scripts/activate.bat
# On Unix-like systems
source .venv/bin/activate

# Sync the MCP server environment (install dependencies, prepare runtime)
uv sync

Next, add the MCP server configuration to your MCP config file. You will reference the exact command and arguments shown here, including the required API key placeholder. The configuration demonstrates how to launch the MCP server via the uv runtime, pointing to the server script Internet_search.py and supplying the BoCha AI API key for access.

{
  "mcpServers": {
    "Internetsearch-mcp-server": {
      "description": "Internetsearch-mcp-server",
      "command": "uv",
      "args": [
        "--directory",
        "/path/Internetsearch-mcp-server",
        "run",
        "Internet_search.py"
      ],
      "env": {
        "BOCHAAI_API_KEY": "sk-123456789412345678312323456789e"
      }
    }
  }
}

Additional content

Security note: The BoCha AI API key shown in the example is a placeholder. Do not use it in production. Obtain a real API key from the BoCha AI service and store it securely in your environment.

Configuration notes: The example uses a local stdio MCP entry powered by uv to run Internet_search.py. The directory flag should point to the actual location of Internetsearch-mcp-server on your system. Ensure your environment variable BOCHAAI_API_KEY is set to a valid BoCha AI API key when you start the server.

Troubleshooting tips: If the server fails to start, verify that the virtual environment is active, the Python dependencies are installed via uv sync, and the Internet_search.py script is accessible at the specified path. Check your BoCha AI API key and network access to the BoCha AI service.