Orenoid

Orenoid / harness-kanban

Public

A cloud-based kanban tool for managing fully containerized coding agents that run 24/7 to handle assigned issues.

agent ai-agents devcontainer harness-engineering kanban
28
0
100% credibility
Found Mar 27, 2026 at 28 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Harness Kanban is a web-based kanban board that deploys isolated AI coding agents to autonomously plan, implement code changes, and create GitHub pull requests for assigned issues.

How It Works

1
📋 Discover Harness Kanban

You hear about a smart board where friendly AI helpers tackle your coding chores around the clock.

2
🛠️ Set it up easily

Download the tool and start it with simple clicks, no tech hassle needed.

3
🔗 Link your workspace

Connect your coding projects so helpers know exactly where to lend a hand.

4
🤖 Hand off a task

Drag a chore to the AI helper on the board and watch it spring into action.

5
Follow the progress
👍
Looks good!

Approve the plan or changes to keep things moving smoothly.

🔄
Tweak it

Give quick feedback so the helper refines and tries again.

🚀 Chores done automatically

Your coding tasks finish themselves with ready-to-review results, freeing you up.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 28 to 28 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 harness-kanban?

Harness-kanban is a cloud-based kanban tool built in TypeScript for orchestrating fully containerized coding agents that run 24/7 to handle assigned GitHub issues. Assign an issue to the code bot, and it spins up isolated Docker containers where AI agents like Codex (Claude soon) plan, code, self-QA via CI/CD checks, and iterate until passing—keeping humans in the loop for approvals and reviews. Deploy via Docker Compose with no local CLI needed, scaling workers across cloud instances.

Why is it gaining traction?

It stands out by containerizing agents to avoid cross-task pollution, enabling safe parallel execution on modest hardware, unlike fragile local AI scripts or bloated cloud-based chatbot projects on GitHub. The async workflow pings you only for human steps, with auto-fixes for failed checks slashing manual debugging. Developers dig the seamless GitHub integration and harness-like wiring for 24/7 agent swarms without vendor lock-in.

Who should use this?

Solo devs or small teams drowning in repo maintenance, like indie hackers automating bug fixes or feature stubs in open-source projects. Perfect for engineering leads piloting AI agents on non-critical issues before scaling to production pipelines, especially if you're eyeing cloud-based project github setups with containerized workflows.

Verdict

Promising MVP for agent-driven dev (28 stars), but 1.0% credibility signals early risks—expect breaking changes and sparse docs. Try on a side repo if you're into experimental harness kanban for coding agents; skip for mission-critical work until more battle-tested.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.