MCP For Claude Marketplace
E2B

E2B MCP Connector for Claude

A+

Secure cloud sandboxes for AI code execution — run Python, JavaScript, and shell commands in isolated Firecracker microVMs with ~150ms cold start.

3 tools Official Updated Oct 1, 2026 Official Vinkius Partner
More Details Connect to Claude

Connect your AI agent to E2B — the leading sandbox platform for AI code execution, trusted by OpenAI, Anthropic, and thousands of AI companies.

What you can do

  • Create Sandboxes — Spin up isolated Linux environments in ~150ms. Each sandbox is a Firecracker microVM with its own kernel, filesystem, and network
  • List Sandboxes — Monitor all active sandbox environments, their templates, and resource usage
  • Kill Sandboxes — Terminate environments when done to release resources and reduce costs

How it works

  1. Subscribe to this server
  2. Enter your E2B API key ($100 free credits to start)
  3. Your agent can now create and manage isolated code execution environments

Who is this for?

  • AI Agent Developers — give agents a safe place to execute generated code without risking your infrastructure
  • Data Science Teams — run analysis scripts in isolated environments with guaranteed reproducibility
  • EdTech Platforms — let students execute code in sandboxed environments with full isolation
sandboxingcode-executionmicro-vmsfirecrackerai-agentsisolated-environment

3 tools expose this connector's capabilities to your AI agent.

list_sandboxes

Useful for monitoring active environments and managing resources. List all currently active sandboxes in your E2B account

create_sandbox

The sandbox is an isolated Linux VM that starts in ~150ms. Use templates like "base" (default), "python3", or "node" for pre-configured environments. Default timeout is 300 seconds. Create a new isolated cloud sandbox for running code securely. Each sandbox is a Firecracker microVM with its own filesystem

kill_sandbox

The sandbox and its filesystem contents are permanently deleted. Terminate a running sandbox by its ID

See how to talk to your AI agent using E2B.

Create a Python sandbox so I can run a data analysis script.

Sandbox created successfully! | Sandbox ID | Template | Status | Info | |---|---|---|---| | `sbx_abc123` | python3 | running | Timeout: 300s | Your sandbox is ready. It's an isolated Linux environment with Python 3.11 pre-installed. Use the sandbox ID to execute code.

Show me all my running sandboxes.

Found 3 active sandboxes: | Sandbox ID | Template | Status | Info | |---|---|---|---| | `sbx_abc123` | python3 | running | Started: 5min ago | | `sbx_def456` | node | running | Started: 12min ago | | `sbx_ghi789` | base | running | Started: 1h ago |

Kill sandbox sbx_ghi789 — I'm done with it.

Sandbox terminated successfully! | Sandbox ID | Status | Info | |---|---|---| | `sbx_ghi789` | killed | Sandbox terminated — resources released |

E2B sandboxes run as dedicated Firecracker microVMs — the same technology used by AWS Lambda and Fargate. Each sandbox has its own Linux kernel, filesystem, and network stack, providing hardware-level isolation. Code running in a sandbox cannot access your host system, other sandboxes, or any external resources unless explicitly configured.

Related Connectors

Aikido Security

Aikido Security MCP

16 tools Official

Query security vulnerabilities via Aikido — list open issues, check repositories, monitor cloud assets, and track compliance directly from any AI agent.

A+ View details →
Filesystem Sandbox Boundary Enforcer

Filesystem Sandbox Boundary Enforcer MCP

3 tools Official

Enforces strict path boundaries and prevents traversal attacks.

A+ View details →
Prisma Cloud

Prisma Cloud MCP

7 tools

Connect Prisma Cloud to any AI agent via MCP.

A+ View details →
Destructive Command Interceptor

Destructive Command Interceptor MCP

3 tools Official

Intercepts and blocks destructive bash commands using pattern matching and obfuscation detection.

A+ View details →