E
E2B Code Sandbox MCP Server
Execute code safely in sandboxed environments from AI agents.
A Model Context Protocol server for E2B. Enables AI agents to execute code in secure, isolated cloud sandboxes with support for multiple programming languages.
Provides safe code execution with file system access, process management, and package installation in ephemeral sandbox environments.
Install
npx -y @e2b/mcp-server MCP Client Config
{
"mcpServers": {
"e2b": {
"command": "npx",
"args": [
"-y",
"@e2b/mcp-server"
],
"env": {
"E2B_API_KEY": "<your-api-key>"
}
}
}
} Required Environment Variables
- E2B_API_KEY
Capabilities
Tools
run_codeinstall_packagesread_filewrite_filelist_sandboxes
Compatible With
Claude Desktop Claude Code Cursor Windsurf Continue
Pricing
Free tier with limited sandbox hours. Paid plans for production use.
Metrics
1,100
GitHub Stars
9,200
Installs
720
Weekly
4
Open Issues
More in AI & Machine Learning
Best of AI & Machine Learning → M
Memory MCP Server
Persistent memory and knowledge graph for AI agent conversations.
open-source 74/100
AI & Machine Learning
S
Sequential Thinking MCP Server
Step-by-step reasoning and problem decomposition for AI agents.
open-source 72/100
AI & Machine Learning
Q
Qdrant MCP Server
Store and search vector embeddings with Qdrant from AI agents.
open-source 70/100
AI & Machine Learning