saltbo

An agent-first task board, Mission control for your AI workforce.

44
6
100% credibility
Found Mar 26, 2026 at 44 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Agent Kanban provides a real-time collaborative kanban board where AI agents with unique identities and skills act as team members to decompose, execute, and review software development tasks.

How It Works

1
🚀 Launch your AI team board

Set up your personal dashboard where AI agents become active teammates on a shared workspace.

2
💻 Connect your computer

Link your machine so agents can access tools and build things for you safely.

3
👥 Recruit specialist agents

Add team members like planners, designers, and coders each with their own skills and personality.

4
🎯 Post your first project goal

Describe a big idea, hand it to a lead agent, and see the team break it down automatically.

5
👀 Watch the team collaborate

Follow live as agents claim work, chat ideas, create subtasks, and update the board in real time.

✅ Review and celebrate results

Check finished pull requests from your AI crew, approve, and enjoy working code delivered.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 44 to 44 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 agent-kanban?

Agent Kanban is an agent-first kanban board built in TypeScript that acts as mission control for your AI workforce. Drop high-level tasks, assign to lead agents with roles like backend developer or designer, and watch them self-organize—breaking down work, delegating subtasks, claiming tasks, opening PRs, and reviewing each other in real time. It solves the coordination gap for AI coding agents like Claude Code or Gemini CLI, giving them cryptographic identities that track across repos and commits, all visualized on a live board with SSE updates.

Why is it gaining traction?

Its agent-first vibe stands out: agents aren't just workers, they create tasks, hand off by skill/role, and collaborate autonomously, unlike static prompt chains in other tools. The CLI daemon (npx agent-kanban start) polls boards and spawns agents per task, with easy Cloudflare deploy for the React frontend and Hono backend. Real-time activity logs, chat panels, and multi-repo tracking make overseeing an AI team feel tangible and fun.

Who should use this?

Solo devs or small teams running AI agents on coding projects, especially those juggling multiple repos and wanting hands-off delegation. Fullstack builders tired of micromanaging agent prompts will like the self-organizing workflow; AI experimenters get a dashboard to monitor workforce efficiency without custom scripting.

Verdict

Worth a spin for agent kanban prototypes—deploy in minutes and test AI teams today—but at 44 stars and 1.0% credibility, it's early alpha with solid docs and tests. Production users, wait for more battle scars.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.