
Code Execution Server
Enables execution of code in a sandbox environment with integrated web search capabilities. Provides a basic framework for running code safely, primarily designed for AI agents and research applications.
README
Code Execution Server
This repository provides a basic implementation of a code execution server, designed primarily for Xmaster (paper, code) and Browse Master (paper code). The full implementation is used in SciMaster.
Due to the proprietary nature of the full code, this repository only includes an open-source framework and the basic components required for code execution. It also includes a simple network search tool implementation.
⚠️ Warning: This is a basic code execution server without virtualization or safety protections. For added security, consider running it within Docker or Apptainer containers as necessary.
🛠️ Setup
Environment
Clone this repository and navigate to the project directory and install the required dependencies:
cd mcp_sandbox/
pip install -r requirements.txt
Tools
- setup the serper key in
configs/web_agent.json
- setup the models' api key in
configs/llm_call.json
🚀 Deploy the Code Execution Server
Step 1: Start the API Server
We will first start the API server used by the tools. This API server proxies all search-related services, including:
- Serper's Google Search Service
- A series of Model APIs
Navigate to the api_proxy directory and start the API server:
cd api_proxy
python api_server.py
Step 2: Deploy the Server
Deploy the server by running the following script in the MCP
directory:
cd MCP
bash deploy_server.sh
📝 Usage
Sending a Request
To send a request to the server, use the following curl
command:
curl -X POST "http://<your-server-url>/execute" \
-H "Content-Type: application/json" \
-d '{"code": "<your code here>"}'
⚡ Benchmarking
For benchmarking, you can run the following command to test the server's performance:
bash benchmarking/pressure.sh 100 100 10 benchmarking/script.lua http://127.0.0.1:30008
Example output:
Running 10s test @ http://127.0.0.1:30008/execute
100 threads and 100 connections
Thread Stats Avg Stdev Max +/- Stdev
Latency 50.21ms 47.15ms 296.96ms 53.20%
Req/Sec 24.13 13.58 130.00 54.99%
23185 requests in 10.10s, 4.27MB read
Requests/sec: 2295.61
Transfer/sec: 432.74KB
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.