What it does
Adds code-interpreter capability by executing code inside an isolated E2B Sandbox (a secure cloud microVM) — for agents needing to run untrusted or generated code (data analysis, scripting, computation) without touching the host. Execution is deterministic and isolated; the LLM writes the code the tool runs. Note: the official e2b-dev/mcp-server repo was archived (deprecated) 2026-04-16 — treat as legacy; the E2B SDK is the current path. Runs locally over stdio.
Example prompts
- Run a Python script to compute summary statistics for this dataset.
- Execute this code snippet in a sandbox and show me the output.
Tools provided
- run_code — Execute code in a secure E2B sandbox and return the output.
What you'll need to connect
This agent will ask you for the following. You enter them when you connect — they're encrypted and never shared with the creator.
- E2B API keyGet this from e2b.dev's account or API settings.Paste the value as a single line.Only sent to: api.e2b.dev
How you're protected
FindAgent runs these safety checks on every agent automatically. They're always on and can't be turned off.
- Prompt-injection scanning
Every request is checked for known prompt-injection and jailbreak attempts before the agent runs. This is always on.
- Secret-leak scanning
Every response is scanned for leaked API keys, tokens, and other secrets before it reaches you. This is always on.
