antopolskiy

antopolskiy / kanban-md

Public

File-based kanban board CLI and TUI for multi-agent workflows. Skills included.

agent-skills ai-agents cli developer-tools go
76
6
100% credibility
Found Feb 10, 2026 at 43 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

kanban-md is a command-line Kanban board tool that stores tasks as plain Markdown files for easy human and AI agent collaboration without databases or servers.

How It Works

1
🔍 Discover simple task tracking

You find a friendly tool to organize your projects like sticky notes on a board, perfect for you and your AI helpers working together.

2
📁 Start your board

In your project folder, you create a new board named after your work, ready to hold all your ideas.

3
✏️ Add your first tasks

You jot down tasks with details like who's doing what, when it's due, and any notes, just like writing on cards.

4
📊 See your board come alive

Open the colorful screen view to watch tasks flow across columns, updating as things change.

5
Work solo or with helpers
👤
You take charge

Drag tasks to the next step with a quick command.

🤖
Assistants join in

Helpers claim tasks automatically, working side by side.

6
✅ Finish and review

Mark tasks done, check progress reports, and celebrate getting organized.

🎉 Projects stay on track

Your work flows smoothly, everyone collaborates without mess-ups, and you always know what's next.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 43 to 76 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 kanban-md?

kanban-md is a file-based kanban board CLI and TUI built in Go as a single static binary. It stores tasks as plain Markdown files with YAML frontmatter, letting you init boards, create tasks, list with filters, move between statuses, edit fields, and pick unclaimed work atomically—all without databases or servers. Ideal for multi-agent workflows, it handles parallel AI agents via claims that expire automatically, plus a watchable TUI for oversight.

Why is it gaining traction?

Unlike SaaS tools or DB-heavy alternatives, kanban-md uses simple files that git seamlessly, with token-efficient compact output for AI prompts and included skills teaching agents kanban workflows. The pick command atomically claims and moves tasks, preventing clashes in multi-agent setups, while commands like board --watch and context generate live summaries or embeddable Markdown. Developers dig the zero-deps runtime and agent-first design over clunky UIs.

Who should use this?

AI workflow builders running multiple agents on code tasks, solo devs ditching Trello for git-native boards, or teams enforcing WIP limits in terminal-heavy environments. Perfect for backend devs scripting task handoffs or frontend folks tracking bugs without SaaS lock-in.

Verdict

Try it if multi-agent kanban fits—docs are thorough, tests extensive, but 33 stars and 1.0% credibility signal early maturity; production use needs more battle-testing. Solid for prototypes.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.