hooosberg

hooosberg / AgentLimb

Public

AgentLimb is a Chrome extension that lets any AI coding tool — Claude Code, Cursor, Codex, Trae, Windsurf, or anything that can run commands — drive your browser. Install the extension, copy one prompt, paste it to your AI, and it auto-configures in 10 seconds.

agentlimb.com
10
1
69% credibility
Found Apr 15, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
HTML
AI Summary

AgentLimb is a Chrome extension that enables AI coding tools to control the browser by saving and replaying action sequences as 'muscles' to reduce token usage on repeat tasks.

Star Growth

See how this repo grew from 10 to 10 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 AgentLimb?

AgentLimb is an HTML-based Chrome extension that lets any AI coding tool—Claude Code, Cursor, Codex, or anything that runs commands—drive your browser directly. Install the extension, copy one prompt from its side panel, paste it into your AI, and it auto-configures in seconds to control your real Chrome sessions with cookies intact. It solves the token waste from AIs re-exploring sites repeatedly by saving action sequences as "muscles" for instant replays.

Why is it gaining traction?

It stands out with one-prompt setup—no API keys or config files—and 16x token savings via muscle memory, skipping redundant exploration on repeat tasks. Developers hook on using their live browser without headless setups or re-logins, plus 100% local operation over HTTP or MCP protocols. The AI-agnostic design means it works with whatever tool you're already using.

Who should use this?

Marketers posting across social platforms daily, researchers scraping and comparing products, ops folks automating form fills and profile updates, or QA engineers testing web apps in real sessions. If you're chaining browser commands in Cursor or Claude Code for repetitive workflows, this cuts the drudgery.

Verdict

Try it for prototyping browser automation—solid multilingual docs and a polished site make onboarding easy, despite just 10 stars signaling early maturity. The 0.699999988079071% credibility score reflects low adoption, but personal use under BSL 1.1 is free and low-risk.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.