The Flaky Test Hunter setup prompt + walkthrough

Paste this into any capable agent and it walks you through the connections, the trigger, and a supervised first run — the whole setup, in your own words. Or skip the setup and run this exact agent hosted.

Set up an agent for me, in its own chat, that does the work of a Flaky Test Hunter. Work from the files and facts I actually give it, rank what matters by impact rather than by how many items it found, and stop short of sending, filing, posting, or changing anything live unless I ask for that in the same run. Walk me through connecting Anthropic. Have it run whenever I hand it the work, and every weekday morning on whatever I left unfinished. Ask me where the source material lives, who should receive the output, and what good enough looks like for a first pass. Run one pass while I watch, then save it.

Already built, running in ~3 min · 30-day money-back on first purchase

#claudecode #github-actions #human-gate #qa #software-factory

About this template

Find the flake. Keep the merge button.

If GitHub Actions keeps going red for reasons that are not real bugs, this template earns its keep in recovered attention.

Flaky tests do two kinds of damage. They burn hours on reruns, and they train your team to ignore red, which is exactly how a real failure slips through. This fork works the problem the way an engineer would: pull the recent failed runs, group them by test name, and score each one flake versus product failure.

Then it drafts a small quarantine or fix PR. You merge. Nothing force-pushes main.

How the flaky test detection works

Fork the template, sign in to Claude Code, and export a GH_TOKEN.
Put your workflow name in ~/vault/ci.md.
Paste: Last 20 failed runs of workflow CI. Report flakes. Draft only.
Read the report. Open the draft PR only if you agree with it.

That is the whole loop. Draft-only by default.

What you get when you fork

The flaky-test-hunter skill, tuned for GitHub Actions failure triage
A vault stub at ~/vault/ci.md where you name your workflow
A demo prompt you can paste in your first five minutes
Your own copy in an isolated Firecracker microVM on your subdomain, still hunting after you close your laptop

What you connect

Bring your own keys. Both stay inside your VM and never leave it:

Claude login or ANTHROPIC_API_KEY
A GH_TOKEN with access to Actions logs and draft PRs

Secrets are scrubbed from the template. You supply your own.

Who it's for

You, if CI red noise is drowning out real bug signal and reruns have become a ritual.

Not you, if you want unattended merges into main. This agent drafts; it never merges.

Fork it

$25+/mo, 30-day money-back, no free trial. Fork it - your copy boots in about 3 minutes, and the first flake report is one pasted prompt away.

Part of the Software Factory collection.

Inside this fork

Forking copies this template into a brand-new, fully isolated microVM on your own subdomain. Here's exactly what lands in it.

Claude Code agent
The upstream harness, pre-installed — same version the creator ran.
Starter VM
1 vCPU · 2 GiB RAM · 10 GiB disk.
Creator's /persist data
The captured persist volume is copied byte-for-byte into your fork.
BYOK — your keys, your VM
Add your model API keys after forking; they live only inside your microVM.

What this agent can do

1 skill

flaky-test-hunter

What you'll configure after forking

Secrets are scrubbed from shared templates — these are the names you supply in your agent's terminal once it boots.

Environment variables

ANTHROPIC_API_KEY

Related templates

Your own Flaky Test Hunter, live in about 3 minutes.

Copy the setup prompt and build it yourself — or let this exact Claude Code agent run hosted, already built, on a brand-new, fully isolated microVM on your own subdomain. Add your own keys after it boots; they never leave the box. Don't love it? Your first jurniti purchase comes with 30 days to get every cent back.

New paid VM · BYOK · 30-day money-back on your first purchase · ~3 min to provision

Flaky Test Hunter

What's inside