Sandboxed code execution for AI agents
Provides secure, isolated sandboxes for AI agents to run code. Popular MCP target for agentic coding workflows.
MCP servers that give your AI agent access to browsers, APIs, repos, and more
AIchitect's Genome scanner detects E2B in your project via these signals:
@e2b/code-interpretere2be2be2b-code-interpreterE2B_API_KEYClaude Code routes code execution to E2B cloud sandboxes via the E2B MCP server, isolating all agent-generated code from the local environment.
→ Safe, isolated execution for AI-generated code — Claude Code can run and iterate on code without any risk to the host machine.
OpenHands runs all agent tasks in E2B cloud sandboxes by default, isolating file writes and command execution from the host.
→ Safe, observable autonomous coding — every agent action happens in a reproducible cloud sandbox with no local environment risk.
Add to your GitHub README
[](https://aichitect.dev/tool/e2b)Explore the full AI landscape
See how E2B fits into the bigger picture — browse all 207 tools and their relationships.