Exa

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": {
    "huazhengwang-exa-search-api": {
      "command": "python",
      "args": [
        "mcp_server/main.py",
        "stdio"
      ],
      "env": {
        "CONFIG": "{\"example_key\":\"example_value\"}",
        "SECURITY": "API_KEY=YOUR_API_KEY",
        "CONFIG_PATH": "path/to/mcp_config.json"
      }
    }
  }
}

You run an MCP Server to connect an MCP client to a Python-based server that processes multi-agent conversations. This guide shows you how to install the server, start it in stdio mode, and configure it for secure operation so you can run and test your MCP-enabled client workflows.

How to use

Start the MCP server in stdio mode to run locally and interact with an MCP client. Use a client to send requests to the server and receive coordinated responses from multiple agents. Configure the server via environment variables to point to a configuration file or JSON string and to set security parameters as needed.

How to install

Prerequisites you need before installation are Python 3.9 or later, Python’s package manager (pip), and the uv tool for optional editable installs.

Step 1: Install the package in editable mode with development dependencies.

pip install -e ".[dev]"

How to install

You can also install dependencies using uv for an editable setup.

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

Running the server

Run the MCP server in stdio mode using the Python entry point. This starts the local server ready to accept MCP client connections.

python mcp_server/main.py stdio

Configuration and security

Configure runtime behavior and security through environment variables. You can provide a JSON configuration string or a path to a configuration file, along with security parameters such as API keys.

Environment variables commonly used:

CONFIG_PATH: Path to a JSON configuration file (for example, mcp_server/mcp_config.json)
CONFIG: A JSON string containing the configuration
SECURITY: Environment variables for security parameters (such as API keys)

Notes on running and testing

The server’s behavior is defined in the main entry point. You can test programmatic interactions by starting the server and using an MCP client to issue requests, ensuring your client is compatible with the server’s MCP interface.

Additional sections

This MCP server is built to work with a defined OpenAPI-based interface and supports local development workflows using Python tooling. You can tailor the configuration to your deployment needs, and you can run quick sanity checks with the provided tooling to ensure your environment is correctly set up before you connect a client.

Available tools

ruff

Lints and formats the Python code to ensure consistency and quality.

mypy

Performs static type checking to catch type errors before runtime.

pytest

Runs unit tests to verify functionality and regressions.

pre-commit

Automates pre-commit checks to enforce coding standards before commits.

hatch

Builds and publishes the MCP project.

uv

A lightweight tool used for editable installs and various development commands.