qarunbook: A test plan your team and your AI can work through

Maker

📌

Hi Product Hunt, I'm Ifeanyi, and I built qarunbook. I've shipped a lot of apps. Every time, before a release, the testing turned into a job of its own: working out what to check, explaining how to check it, and chasing up the bugs. The one that hurt most was a bug marked "fixed" that nobody looked at again on the new build. qarunbook keeps that work in one place: - A plan that says how to test. Each check has steps and an expected result. You can write the plan, import it from Markdown, CSV or Excel, or have your AI draft it with the free /test-checklist skill. - Results on every platform, so web, Android and iOS each get their own pass or fail. - Issues stay on the check that found them. When an issue is marked fixed, the check shows "retest", not "passed", until someone looks again. - Your AI on the same plan. Claude Code, Cursor, Codex or any MCP client can read what failed, record results and follow up on fixes, and it acts under your name and role. - Bug tracking after launch. Once an app is live, you can switch it to tracking issues, with or without a check behind each one. To be clear about what it isn't: qarunbook doesn't run automated tests. It's for the testing that people (and their AI assistants) do on a real build. It's free for 2 apps and 5 people, with no card. Paid plans are a flat price per workspace, never per seat. I'd love your feedback. If you already have a test plan somewhere, try importing it, then tell me where it didn't fit the way your team tests.

Report

1d ago

Is there a way to assign specific tests to different team members?

Report

2h ago

How much control does the team have over what the AI can change or mark as complete?

Report

3h ago

Are there filters or views for finding failed tests and pending retests quickly?

Report

1h ago