XMihura

XMihura / Kanvas

Public

Manage projects visually with humans and AI agents using Obsidian Canvas.

72
9
100% credibility
Found Mar 14, 2026 at 70 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Kanvas is a workflow tool for Obsidian Canvas boards that uses colors and arrows to plan, track, and coordinate projects collaboratively between people and AI agents.

How It Works

1
🗺️ Discover Kanvas

You learn about a simple way to plan and track projects visually on a canvas board in your note-taking app, working side by side with AI helpers.

2
📥 Set up your board

Copy a few helper files and a starting board into your project folder, then open it in your note app to see the colorful legend.

3
🎨 Map out your project

Draw boxes for project areas, add task cards with meanings like 'to do' in red or 'done' in green, and connect them with arrows to show what waits on what—your plan comes alive!

4
🤝 Team up with AI

Share the rules with your AI helper so it can suggest new ideas, start tasks, or mark progress safely, while you review and approve changes on the board.

5
🔄 Track and update

Watch tasks shift colors as they move from planning to doing to finished, with blocked ones automatically pausing until ready—it feels organized and smooth.

6
🔧 Add auto-checks (optional)

Install a little add-on that spots issues and keeps the board tidy while you edit visually.

🏆 Project complete!

Celebrate as your entire project wraps up with a beautiful visual story of every step, saved right with your work for easy review anytime.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 70 to 72 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 Kanvas?

Kanvas turns Obsidian Canvas into a visual board for managing projects between humans and AI agents, using color-coded cards for states like proposed (purple), to-do (red), doing (orange), review (cyan), and done (green). Arrows track dependencies, with blocked tasks auto-graying out, solving the gap where AI excels in CLI sandboxes but needs structured planning and progress tracking. A Python CLI enforces rules—no direct JSON edits—with commands like `status`, `ready`, `start ID`, `propose group "title" "desc"`, keeping agents honest across tools like Claude Code or GitHub Copilot.

Why is it gaining traction?

It stands out as agent-agnostic and Git-friendly: Canvas JSON files diff and merge naturally in repos, unlike rigid Jira boards or siloed AI prompts. Low setup—just Python 3.7+ and Obsidian—beats bloated PM tools, with optional JS watcher for real-time linting of cycles and warnings. Developers hook on the shared visual space that evolves with commits, blending manage projects in Obsidian seamlessly with AI execution.

Who should use this?

Indie devs or small teams coordinating AI agents for code tasks, like frontend engineers breaking down features or ops folks managing GitHub Copilot licenses and subscriptions visually. Ideal for those ditching Jira for lighter manage projects in Obsidian flows, or hybrid human-AI workflows in GitHub orgs without Terraform overhead.

Verdict

Promising for AI-augmented project management, with excellent docs and examples, but early maturity at 40 stars and 1.0% credibility score means test in side projects first. Grab it if you're experimenting with agents in Obsidian—MIT license makes it risk-free.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.