Code Execution

Provides a minimal code execution MCP server with a simple API proxy and deployment flow for testing and prototyping.

python

13

GitHub Stars

python

Language

6 months ago

First Indexed

2 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": {
    "sjtu-sai-agents-mcp_sandbox": {
      "command": "python",
      "args": [
        "api_server.py"
      ]
    }
  }
}

You can run a basic code execution server designed to support lightweight code execution workflows. It provides a simple API and a deployment flow to bring the server up locally for testing, experiments, or integration demonstrations. This setup includes a minimal API proxy for search/model services and a deployment script to start the MCP server in a development environment. Note that this is a basic implementation without full virtualization or safety protections, so it’s advised to use containers if you need stronger isolation.

How to use

Send code to the server through the standard execute endpoint exposed by the MCP server. You submit your code payload to the server, and it runs the code and returns the results. You can benchmark performance to understand latency and throughput under your expected load. Interactions are designed for practical experimentation and local testing with an MCP client.

You can trigger execution by sending a POST request to the execution endpoint with your code. The server is designed for straightforward usage patterns suitable for rapid iteration and prototyping.

How to install

Prerequisites: you need Python and a functioning Python package environment. You also require a shell to run the setup and deployment scripts.

Install Python dependencies

pip install -r requirements.txt

Prepare environment for the API proxy and MCP server deployment
Start the API proxy server

cd api_proxy
python api_server.py

Deploy the MCP server

cd MCP
bash deploy_server.sh

Security and notes

This is a basic code execution server without virtualization or safety protections. For added security, consider running it within containment such as Docker or Apptainer containers as appropriate for your workload.

Available tools

network_search

A simple network search tool integrated with the server to proxy search-related services and model APIs.