e2b-cookbook

Why we included this project

This is a hands-on collection of runnable examples for building agents and code interpreters inside E2B's sandboxed execution environments, so it works well as a learn-by-doing resource. The cookbook matches each major LLM provider, including OpenAI, Anthropic, Mistral, Groq, Fireworks, Together, and WatsonX, with working Python and TypeScript snippets for data analysis, code interpretation, and agentic workflows. It also covers integrations with frameworks like LangChain, LangGraph, CrewAI, AutoGen, and the Vercel AI SDK, plus recipes for running Docker, Playwright, and MCP servers inside a sandbox. For a developer evaluating secure code execution for an agent, the examples read as ready-made starting points rather than abstract documentation. It is not deployable software itself, but a curated reference whose value is the breadth of copy-paste patterns across providers and tooling.