• Picks
  • Discover
  • Alternatives
  • About

© 2026 Willet

  • Newsletter
  • About
  • FAQ
  • Terms
  • Privacy & Cookies
Join the Discord (opens in a new tab)
Pydantic avatar

Monty

Python interpreter in Rust for safely running LLM-generated code

  • Agent frameworks and tools
  • Languages and frameworks
View on GitHub
Website

What it is

Monty is a sandboxed Python interpreter written in Rust that executes code produced by language models without spinning up containers. The sandbox has no filesystem, environment variables, or network; it can only reach the host through functions and mounts you explicitly provide, and it supports resource limits and snapshots. Developers building AI agents embed it from Python, JavaScript, or Rust to run model-written code with sub-millisecond startup.

Images

Facts

Stars
8,740 GitHub stars.
8,740

Forks
466
Commits
874
Contributors
65
Language
Rust
License
MIT
Last commit
Oct 20, 2026
On GitHub since
May 2023
First release
January 2026
Platforms
library
Self-hostable
No
Tags
  • agent-sandbox
  • ai-security
  • ai-agents

Last synced from GitHub on Oct 20, 2026

The 10 best overlooked open-source projects, every Monday.

Share and more

Comments

No comments yet. Start the conversation.

Similar projects

  • opensandbox-group avatar

    OpenSandbox

    Isolated sandbox runtime for AI agents on Docker or Kubernetes

    16k
  • Microsoft avatar

    MXC

    Cross-platform sandbox SDK for running untrusted code and model output

    2.9k
  • NVIDIA Corporation avatar

    OpenShell

    Policy-enforced sandbox runtime for running autonomous AI agents

    16.8k