✴️ E2B Cookbook

Example code and guides for building with E2B SDK.

Read more about E2B on the E2B website and the official E2B documentation.

Examples

Hello World guide

• TypeScript
• Python

Open-source apps

• E2B AI Analyst - analyze your data & create interactive charts
• E2B Fragments - prompt different LLMS to generate apps with UI
• E2B Surf - computer use AI agent powered by OpenAI

LLM providers

Provider

Topic(s)

Example

Python

TypeScript

OpenAI

Agents SDK

Agentic workflows running in E2B sandboxes

Python

Agents API

Full-stack workbench, one E2B sandbox per chat

Python

Agents API

Webhook-managed sandbox lifecycle per session

Python

GPT-5.6

Data analysis and visualization of a CSV

Python

TypeScript

GPT-5.6

Code interpreter and reasoning on image data

Python

TypeScript

GPT-5.6

Code interpreter for ML on dataset

Python

TypeScript

Codex CLI

OpenAI Codex, running inside a Sandbox

Python

TypeScript

Anthropic

Claude Opus 5

Code interpreter

Python

TypeScript

Claude Code

Claude Code, running inside a Sandbox

Python

TypeScript

Claude Managed Agents

Self-hosted worker running inside a Sandbox

Python

Meta

Muse Spark

Coding agent with E2B as its execution backend

Python

Mistral

Codestral

Code interpreter

Python

TypeScript

Groq

Llama 3

Code interpreter via function calling

Python

TypeScript

Fireworks AI

Qwen2.5-Coder-32B-Instruct

Code interpreter

Python

Llama 3.1 405B, 70B, 8B

Code interpreter

Python

Together AI

Llama 3.1, Qwen 2, Code Llama, DeepSeek Coder

Code interpreter

Python

TypeScript

WatsonX AI

IBM Graphite, Llama, Mistral

Code interpreter

Python

TypeScript

AI frameworks integrations

Framework

Description

Python

TypeScript

🦜⛓️ LangChain

LangChain with Code Interpreter

Python

🦜🕸️ LangGraph

LangGraph with code interpreter

Python

CrewAI

CrewAI agent with sandboxed Python execution

Python

Hermes Agent

Learn an incident-triage skill in one sandbox and apply it in a fresh session

Python

▲ Vercel AI SDK

Next.js + AI SDK + Code Interpreter

TypeScript

▲ Vercel AI SDK

AI SDK sandbox provider: sandboxed tools via restricted sessions, and harness coding agents (Claude Code, Codex) running inside E2B

TypeScript

▲ Vercel eve

Feedback analyst agent whose sandbox backend is E2B, publishing an HTML report from the sandbox

TypeScript

Flue

Feedback analyst agent running entirely inside an E2B sandbox, publishing an HTML report from the sandbox

TypeScript

Pi

Pi builds its own E2B code-interpreter extension in a sandbox, then uses it to analyze data

TypeScript

AgentKit

AgentKit Coding Agent

TypeScript

Sandbox Agent SDK

Run Sandbox Agent inside E2B and connect with the SDK

TypeScript

Stirrup

The lightweight framework for building agents

Python

Remote execution integrations

Integration

Description

Python

TypeScript

Crabbox

Warm one E2B sandbox, sync local changes, and rerun a test suite on the same lease

Python

Model Context Protocol (MCP)

Example

Description

TypeScript

MCP Client

Basic MCP client connection to E2B sandbox

TypeScript

MCP Custom Server

Connect to custom filesystem MCP server from GitHub

TypeScript

MCP Custom Template

Create custom E2B template with pre-installed MCP servers

TypeScript

MCP Research Agent

Research agent using arXiv and DuckDuckGo MCP servers

TypeScript

MCP Claude Code

Claude Code with MCP integration

TypeScript

MCP Browserbase

Web automation agent using Browserbase MCP server

TypeScript

MCP Groq Exa (deprecated)

AI research using Groq with Exa MCP server

TypeScript

Example use cases

• Upload dataset and analyze it with Llama 3.3 - Python
• Scrape Airbnb and analyze data with Claude Opus 5 and Firecrawl - TypeScript
• Visualize website topics with Claude Sonnet 5 and Firecrawl - Python
• Next.js app with LLM + Code Interpreter and streaming - TypeScript
• How to run a Docker container in E2B - Python/TypeScript
• Tailcat: encrypted links between sandboxes, and between a sandbox and your laptop - Python/TypeScript
• How to run Playwright in E2B - TypeScript
• Map custom subdomains to your sandboxes - TypeScript
• Delete old paused sandboxes by age and metadata - Python/Shell
• Feedback analyst agent on Flue, publishing an HTML report from a sandbox - TypeScript
• Warm an E2B sandbox and rerun tests after adding a local regression case - Python
• Teach Hermes an incident-triage playbook and reuse it in a fresh session - Python

Running the examples as a test suite

Every example is exercised nightly against live E2B by tests/run-examples.ts: each one is uploaded into a fresh sandbox, installed with its own toolchain (npm, uv, Poetry, or nbconvert for notebooks) and run. An example passes if it exits 0.

npm install
npm test                      # all of them
npm test -- hello-world       # substring filter, one or a few

You need an E2B_API_KEY plus whichever provider key the examples you are running use - see .env.example. hello-world-js and hello-world-python need only the E2B key, so they are the ones to try first.

What counts as a failure. The suite checks the sandbox, not the model. A provider rate limit or an exhausted quota counts as OK, because the sandbox still built, installed and ran; non-deterministic model behaviour - no chart produced, a malformed tool call, generated code raising inside the sandbox - is reported as skipped. Real failures are the things that are actually broken: missing templates, dependency resolution, wrong entrypoints, auth, retired model ids, sandbox timeouts.

Some examples are deliberately not covered - they need a provider key this repo does not hold, a long-lived server the runner cannot assert on, or an upstream fix. Each exclusion is listed with its reason at the top of tests/run-examples.ts, so a gap is never silent.