Custom Search

MCP Server generated by mcp.ag2.ai

python

0

GitHub Stars

python

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "qingyun-wu-custom-search-api": {
      "command": "python",
      "args": [
        "mcp_server/main.py",
        "stdio"
      ],
      "env": {
        "CONFIG": "YOUR_CONFIG_JSON_STRING",
        "SECURITY": "YOUR_SECURITY_PARAMETERS",
        "CONFIG_PATH": "YOUR_CONFIG_PATH"
      }
    }
  }
}

You can run an MCP server to expose an OpenAPI-driven MCP endpoint for the Google Custom Search API. This server lets clients connect via standard MCP transports and negotiate conversations with agents, using a prebuilt configuration generated from the specified OpenAPI spec. It’s practical when you want a dedicated MCP service that orchestrates multi-agent interactions around the Custom Search API.

How to use

You interact with the MCP server through an MCP client. Start the server using the provided stdio transport, then connect your client to the local process. The server supports multiple transport methods, including stdio, server-sent events (sse), and streamable HTTP pipelines. Once running, your client can request search-related actions defined by the underlying OpenAPI-driven MCP configuration and receive coordinated responses from participating agents.

How to install

Prerequisites: Python 3.9 or newer, pip, and uv.

Clone the project directory and navigate into it.

git clone <repository-url>
cd mcp-server

Install dependencies using the editable dev install option. You can run the standard pip workflow or use uv to manage editable installs.

pip install -e ".[dev]"

uv pip install --editable ".[dev]"

Notes on running the server

To start the MCP server, run the stdio version of the main entry point. This launches a local process that your MCP client can attach to via standard input/output.

python mcp_server/main.py stdio

Configuration and security

Configure the server using environment variables. The following options are supported: CONFIG_PATH for the path to a JSON configuration file, CONFIG for a JSON string containing the configuration, and SECURITY for security-related parameters (such as API keys). These controls influence how the server loads settings and enforces access.

Building, testing, and quality checks

The project uses linting, static analysis, and tests to maintain quality. Run lint checks, format, static analysis, and tests as part of your development workflow to keep the codebase healthy.

Troubleshooting and notes

If you encounter issues starting the server, verify that Python 3.9+ is in use, dependencies are installed, and the environment variables are set correctly. Inspect error messages for guidance on missing config values or bad JSON in CONFIG_PATH. For transport-related problems, ensure the chosen MCP transport (stdio, sse, or streamable-http) is supported by your client and that the server process is reachable.

Examples and tips

Use the stdio transport during local development to keep a tight feedback loop with your MCP client. When deploying, consider switching to a more scalable transport such as streamable-http if your workflow requires remote clients to connect over HTTP.

Available tools

lint

Run lints and format checks to ensure code quality (ruff) and formatting (ruff format).

static-analysis

Perform static analysis including type checks (mypy) and additional security scans (bandit, semgrep).

tests

Execute unit and integration tests with coverage reporting.