faveos8758

TypeScript sandbox for running untrusted agent code: V8 isolates (isolated-vm), Docker option, Python/bash subprocesses, Express API

20
0
100% credibility
Found Apr 19, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A library and web service for safely executing untrusted code in JavaScript, TypeScript, Python, or Bash within isolated environments.

How It Works

1
🕵️ Discover the safe code runner

You hear about a special playground that lets AI helpers run code snippets without risking your computer.

2
📦 Get it on your computer

You download the tool and set it up quickly so it's ready to use.

3
🛡️ Create your protected space

You make a secure area where code can run, setting simple limits like time allowed and space it can use to stay safe.

4
🚀 Feed it code from your AI

You paste in the code your AI created, pick the language like JavaScript or Python, and watch it run safely.

5
📊 See the safe results

You get back exactly what the code produced, like printed numbers or messages, with no danger to your files or system.

🎉 AI code runs securely

Now your AI can experiment with code anytime, and everything stays protected and under control.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ts-agent-sandbox?

This TypeScript library provides a secure sandbox for running untrusted JavaScript, TypeScript, Python, or Bash code in Node.js environments, perfect for AI agent execution where code safety matters. Developers get timeouts, memory limits, optional network access, and virtual file systems via a simple API or Express HTTP endpoint at POST /execute. It supports quick npm installs for typescript code sandbox setups, with Docker for extra isolation on JS/TS runs.

Why is it gaining traction?

It stands out with multi-language support in one package—no switching tools for JS/TS agents versus Python scripts—plus configurable options like allowlisted fetch hosts and tmpdir file access that users control directly. The HTTP API and streaming results make it drop-in ready for typescript github actions or agent APIs, beating basic eval hacks or heavy alternatives. Early adopters like it for node typescript sandbox reliability without full VM overhead.

Who should use this?

Backend engineers building AI agent platforms or dynamic code eval in Node apps, such as typescript github copilot integrations or reactjs typescript sandbox testers. It's for devs handling untrusted inputs in typescript github api services, github issues automation, or typescript playground sandbox online demos—especially those needing python/bash alongside JS without separate services.

Verdict

With 20 stars and 1.0% credibility score, it's early-stage but promising: solid docs, full test suite, and Docker images make it usable now for prototypes. Grab it via typescript sandbox npm if you need a lightweight agent sandbox today, but watch for community growth before production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.