Before and after
A real diff in the Express repository: a freshness check extended to a new HTTP method, and a header guard in res.send, with the tests as they were before the change.
After (real output, excerpt):
1 | req.fresh can return true for QUERY (lib/request.js:472) | Not covered | test/req.fresh.js and test/req.stale.js only send GET requests.
4 | res.send skips Content-Length when Transfer-Encoding is set | Not covered | res.send.js:270 and :287 set Transfer-Encoding, but only with status 204 or 205.
Whether QUERY freshness is intended to follow the GET semantics is not stated in the diff.
See it run

What it does
Give it a diff and the project's tests. You get the behaviours the change adds or alters, each marked Covered, Not covered or Unknown against the tests, and proposed test cases for the gaps with inputs and expected results taken from the code. It never gives a coverage percentage and never says a test passes unless it ran.
The evidence
1 real-data run on a public multi-hunk change in the Express repository (Claude). We checked each Covered and Not covered claim against the tests ourselves.
More on how it was tested
We checked each Covered and Not covered claim by searching the tests ourselves: all matched, and the proposed test for the Transfer-Encoding case is realistic. Not yet run on other models.
Limits
- It reads tests; it does not run them.
- Not covered means it searched and found nothing.
- No coverage percentages.
Questions
Questions or a problem?
Which assistants does it work with?
Works in ChatGPT or any chatbot: you paste in your details and it writes the result. A repo with the tests, or the diff and test files pasted.
For technical users
Using it with an AI assistant
- Run
npx skills add proskillpacks/skillsin your project and pick this skill. Or copy thetest-gap-finderfolder from GitHub (https://github.com/proskillpacks/skills/tree/main/skills/developers/test-gap-finder). - Put the
test-gap-finderfolder where your assistant looks for skills. The generic path is.agents/skills/(ornpx skills addfrom a folder). Tool-specific paths:- Claude Code:
~/.claude/skills/(or.claude/skills/inside one project) - OpenAI Codex, Gemini CLI, Cursor and others:
.agents/skills/or the skills folder their docs name - claude.ai: zip the skill folder yourself, then Customize, Skills, the plus button, Create skill, Upload a skill (code execution must be on in Settings, Capabilities)
- ChatGPT or any chatbot: open the skill's
prompt.md, paste it into the chat, then paste your own text or link - Steps for each tool, with links to its official docs: how to install skills
- Claude Code:
- Ask for the job in plain words, for example: "What tests am I missing for this change? The diff is in change.diff.". The skill loads by itself.
Jobs that go with this
- Write a changelog from your commits (Free): A list of commits in, release notes people can read.
- Write a pull request description and review notes (Free): Your branch in, a pull request write-up and test plan.
- Browse the web with a local Qwen model, fewer wrong turns (Free): A browsing task and your local model in, the task done in fewer tries.