Claude Prompt Library

30 Claude Code Prompts for Tests

30 copy-paste prompts

Name the files and the test command. Claude Code writes or fixes tests in that scope, does not wander, and tells you what it ran. Not a coverage vanity pass and not a rewrite of the app.

In short: This page contains 30 copy-paste ready prompts, organized into 6 categories with a description and pro tip for each. The first 5 prompts are free instantly, no signup needed. Hand-curated and tested by the AI Academy team.

By Louis Corneloup · Founder, Techpresso

Last updated September 8, 2026·Hand-curated & tested by the AI Academy team

Add

5 prompts

Regression

1/30

✨ What it does

Claude Code adds a regression test in [BUG] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a test writer working in Claude Code. <context> I need you to add a regression test for this bug. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [BUG] - Files: [FILES] - Test command: [TEST] </inputs> <task> 1. Test the bug in [FILES] 2. Run [TEST] 3. Then fix if I also asked </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a runner.

Unit

2/30

✨ What it does

Claude Code adds unit tests in [CODE] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a unit writer working in Claude Code. <context> I need you to add unit tests in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [CODE] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Tests in [FILES] 2. Run [TEST] 3. Same public API </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent cases I forbade.

Table

3/30

✨ What it does

Claude Code adds table-driven tests in [CASES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a table writer working in Claude Code. <context> I need you to add table-driven tests in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [CASES] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Cases from [CASES] only 2. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a case.

Edge

4/30

✨ What it does

Claude Code adds the edge tests in [EDGES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are an edge writer working in Claude Code. <context> I need you to add edge tests I named. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [EDGES] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Only [EDGES] 2. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent leap-year.

Golden

5/30

✨ What it does

Claude Code adds a golden test in [OUTPUT] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a golden writer working in Claude Code. <context> I need you to add a golden test I can update. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [OUTPUT] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Golden in [FILES] 2. Run [TEST] 3. How to update </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a snapshot I cannot see.

Fix

5 prompts

Failing

6/30

✨ What it does

Claude Code fixes the failing test in [ERROR] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a test fixer working in Claude Code. <context> I need you to fix this failing test in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [ERROR] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Fix in [FILES] 2. Run [TEST] 3. Prefer fix the test if the app is right, or say which </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not silence the assert.

Flake

7/30

✨ What it does

Claude Code deflakes the test in [TEST NAME] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a flake fixer working in Claude Code. <context> I need you to deflake this test in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [TEST NAME] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Deflake in [FILES] 2. No sleep unless I allow 3. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent CI minutes.

Slow

8/30

✨ What it does

Claude Code speeds the test in [TEST NAME] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a slow-test fixer working in Claude Code. <context> I need you to speed this test in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [TEST NAME] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Speed in [FILES] 2. Keep cases 3. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent 200ms.

Brittle

9/30

✨ What it does

Claude Code unbrittles the test in [TEST NAME] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a brittle fixer working in Claude Code. <context> I need you to unbrittle this test in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [TEST NAME] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Assert behavior not internals I forbade 2. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a class name.

Delete Dead

10/30

✨ What it does

Claude Code deletes dead tests in [NAMES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a dead-test cutter working in Claude Code. <context> I need you to delete dead tests in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [NAMES] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Delete only [NAMES] if dead 2. Run [TEST] 3. Ask if unsure </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not delete a case I still need.

Kinds

5 prompts

API

11/30

✨ What it does

Claude Code adds API tests in [ENDPOINTS I HAVE] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are an api tester working in Claude Code. <context> I need you to add API tests in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [ENDPOINTS I HAVE] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Only those endpoints 2. Run [TEST] 3. No invented route </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent /v2.

UI

12/30

✨ What it does

Claude Code adds UI tests in [FLOWS I HAVE] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a ui tester working in Claude Code. <context> I need you to add UI tests in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FLOWS I HAVE] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Only [FLOWS I HAVE] 2. Run [TEST] 3. No invented selector soup if I gave roles </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a testid.

SQL

13/30

✨ What it does

Claude Code adds SQL tests in [QUERIES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a sql tester working in Claude Code. <context> I need you to add SQL tests in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [QUERIES] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Tests for [QUERIES] 2. Run [TEST] 3. No invented fixture rows unless I gave them </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a user.

Auth

14/30

✨ What it does

Claude Code adds auth tests in [CASES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are an auth tester working in Claude Code. <context> I need you to add auth tests in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [CASES] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Cases from [CASES] 2. Run [TEST] 3. No real secrets </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a JWT.

Script

15/30

✨ What it does

Claude Code adds script tests in [SCRIPT] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a script tester working in Claude Code. <context> I need you to add tests for this script. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [SCRIPT] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Tests in [FILES] 2. Run [TEST] 3. Keep flags I named </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a flag.

These prompts give you the what. Tutorials give you the why.

Learn when to use extended thinking, how to build Claude Projects, and workflows that compound. 300+ tutorials and growing.

Try AI Academy Free

Scope

5 prompts

One File

16/30

✨ What it does

Claude Code tests one file in [FILE] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a one-file tester working in Claude Code. <context> I need you to tests in one file only. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILE] - Test: [TEST] </inputs> <task> 1. Only [FILE] 2. Run [TEST] 3. Ask if you need more </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not open neighbors.

No New Runner

17/30

✨ What it does

Claude Code uses my test runner in [FILES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a runner tester working in Claude Code. <context> I need you to use my test runner only. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Test: [TEST] </inputs> <task> 1. Tests in [FILES] 2. Run [TEST] 3. No new framework </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent Playwright if I use Vitest.

No Snapshot Spam

18/30

✨ What it does

Claude Code avoids huge snapshots in [FILES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a snapshot tester working in Claude Code. <context> I need you to avoid huge snapshots unless I ask. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Test: [TEST] </inputs> <task> 1. Prefer explicit asserts 2. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a 400-line snapshot.

Coverage Honest

19/30

✨ What it does

Claude Code covers the named paths in [PATHS] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a coverage tester working in Claude Code. <context> I need you to cover what I named only. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [PATHS] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Cover [PATHS] 2. Run [TEST] 3. No fake 100% </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent coverage.

Fixture Honest

20/30

✨ What it does

Claude Code uses existing fixtures in [FIXTURES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a fixture tester working in Claude Code. <context> I need you to use fixtures I have only. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FIXTURES] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Only [FIXTURES] 2. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a user.json.

After

5 prompts

Run And Paste

21/30

✨ What it does

Claude Code runs tests and pastes output in [TEST] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a runner working in Claude Code. <context> I need you to run tests and paste the result. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [TEST] - Files: [FILES] </inputs> <task> 1. Run [TEST] 2. Paste output 3. Fix only if I asked </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a pass.

Watch Ask

22/30

✨ What it does

Claude Code runs tests once in [TEST] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a watch tester working in Claude Code. <context> I need you to do not start a watcher unless I ask. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [TEST] - Files: [FILES] </inputs> <task> 1. One run of [TEST] 2. No watch </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Watchers hang the session.

CI Note

23/30

✨ What it does

Claude Code writes the CI test note in [TEST] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a ci tester working in Claude Code. <context> I need you to write how CI should run this. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [TEST] - Files: [FILES] - CI I have: [CI OR VERIFY] </inputs> <task> 1. Note from [CI] or ask 2. Stay about [FILES] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent GitHub Actions.

Fail Fast

24/30

✨ What it does

Claude Code runs fail-fast in [TEST] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a fail-fast tester working in Claude Code. <context> I need you to stop after first fail if I said so. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [TEST] - Files: [FILES] </inputs> <task> 1. Run fail-fast if the runner can 2. Paste 3. Fix in [FILES] if I asked </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not invent a flag.

Update Snapshot

25/30

✨ What it does

Claude Code updates snapshots on GO in [TEST] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a snapshot updater working in Claude Code. <context> I need you to update snapshots only if I said. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [TEST] - Files: [FILES] </inputs> <task> 1. Update only if I said GO 2. Stay in [FILES] 3. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not -u by default.

Most people use 10% of Claude. Tutorials unlock the rest.

AI Academy: 300+ hands-on tutorials on Claude, ChatGPT, Midjourney, and 50+ AI tools. New tutorials added every week.

Start Your Free Trial

Fixes

5 prompts

Undo Wander

26/30

✨ What it does

Claude Code reverts extra test edits in [DIFF] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a scope tester working in Claude Code. <context> I need you to revert out-of-scope test edits. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [DIFF] - Allowed: [FILES] </inputs> <task> 1. Revert extras 2. Keep the tests </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Neighbors stay.

Tests Plus Fix

27/30

✨ What it does

Claude Code writes the test then the fix in [BUG] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a combo tester working in Claude Code. <context> I need you to failing test then the fix. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [BUG] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Red 2. Fix 3. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Red first if I asked.

Tests Plus Note

28/30

✨ What it does

Claude Code adds tests plus a PR note in [FILES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a combo tester working in Claude Code. <context> I need you to tests plus a PR note. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Test: [TEST] </inputs> <task> 1. Tests 2. Note </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Note lists what is covered.

Cut Weak

29/30

✨ What it does

Claude Code cuts the weak tests in [WEAK] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a weak-test cutter working in Claude Code. <context> I need you to cut weak tests I named. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [WEAK] - Files: [FILES] - Test: [TEST] </inputs> <task> 1. Cut [WEAK] 2. Run [TEST] </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Do not cut a strong case.

Name Tests

30/30

✨ What it does

Claude Code renames tests in [FILES] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a names tester working in Claude Code. <context> I need you to rename tests to say the behavior. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Test: [TEST] </inputs> <task> 1. Rename in [FILES] 2. Run [TEST] 3. No extra cases </task> <constraints> - Plan before you edit if more than two files change. - Do not touch files I mark OFF LIMITS. - Run the test command I gave. If I gave none, ask before inventing one. - Do not invent APIs, env vars, or tickets. </constraints> <format> A short plan, then the edits, then what you ran and what I should verify. </format>

💡

Pro tip: Names should fail in English.

Free tool

Prompt Optimizer

Turn a rough idea into a structured, professional AI prompt.

Try it free →

Frequently Asked Questions

They are written for Claude Code in the terminal: plan, edit, test, and stay inside tests. You can paste a shorter version into claude.ai if you are not in a repo.

Each prompt sets scope and non-goals. Still use Plan mode on multi-file work, and read the diff before you accept edits.

It helps. A short CLAUDE.md with stack, test command, and files Claude must not touch keeps these prompts from inventing a second style guide.

The prompts are free to copy. Claude Code needs your Anthropic login and uses your plan's usage. Local files never leave your machine except through that session.

The hub is a 21-prompt overview. This page is only tests: thirty copy-paste jobs with scope, verify, and non-goals already written.

Prompts are the starting line. Tutorials are the finish.

A growing library of 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ AI tools. New tutorials added every week.

Start 7-Day Free Trial Explore AI for Entrepreneurs

7-day free trial. Cancel anytime.

Browse All Claude Prompt Collections Browse All Prompt Collections

Popular prompt collections

AI Photo Editing Prompts ChatGPT Jailbreak Prompt Best ChatGPT Prompts Funny Hinge Prompts ChatGPT Caricature Prompts