agent-bridges

agent-bridges / ab

Public

agent-bridge

agent-bridge.dev agent-bridge ai-agents canvas devtools open-source
10
0
89% credibility
Found Mar 25, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Shell
AI Summary

Agent-Bridge provides a persistent online workspace for managing AI coding agents with customizable canvases, shared terminals, and file access.

How It Works

1
🔍 Discover Agent-Bridge

You find this helpful tool for creating shared workspaces where AI helpers can code together persistently.

2
⚡ Start quick demo

With one simple command, you launch a ready-to-use demo that sets everything up on your computer.

3
🚀 Open your workspace

Your personal canvas dashboard appears in your web browser, feeling fresh and ready to explore.

4
🔐 Log in easily

Use the default name and password to enter, and you're inside your new AI coding space.

5
🎨 Browse the canvas

Drag windows around, open notes, files, and terminals tailored for each AI helper.

6
💻 Connect AI agents

Add your AI coding buddies, watch them get their own persistent terminals and tools.

✅ Enjoy lasting sessions

Your AI agents now work in a saved, collaborative environment that picks up right where you left off.

Sign up to see the full architecture

5 more

Sign Up Free

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 ab?

Agent-Bridge (ab) is a Docker-orchestrated platform for persistent workspaces tailored to AI coding agents, combining PTY terminals, a backend API for agent management, and a canvas frontend for layouts and components. It tackles flaky agent sessions by saving per-agent states like viewports, windows, files, and notes, all accessible via http://localhost:5281 after a one-command quick-start. Shell scripts and Compose files handle prod, dev, and test stacks, pulling in remote agents like docker-test-agent for instant demos.

Why is it gaining traction?

Its hook is seamless persistence across PTY sessions and canvases with draggable windows, minimaps, rulers, and file actions—features absent in basic agent-bridge setups. Quick curl install or scripts/stack.sh up skips repo cloning for prod deploys, while dev bind-mounts enable live changes. Test stacks with pre-seeded agents make validation dead simple, standing out from fragmented ab testing frameworks on GitHub.

Who should use this?

AI devs building terminal agents for Claude or Codex, needing shared canvases for multi-agent coordination. Teams prototyping persistent envs for LLM coding hooks, avoiding session resets. Ops folks deploying agent-bridge stacks via SSH daemon pushes.

Verdict

Solid docs, demo video, and modes like quick-start make ab usable now, despite 10 stars and 0.9% credibility score signaling early maturity (v0.1.6). Try the curl installer for agent-bridge proofs-of-concept; skip for production without child repo audits.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.