Claude Prompt Library

30 Claude Code Prompts for Refactors

30 copy-paste prompts

Name the files and the rule. Claude Code refactors inside that scope, keeps behavior unless you allow a change, and tells you what it ran. Not a rewrite and not a framework swap.

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

Small

5 prompts

Extract Function

1/30

✨ What it does

Claude Code extracts a function 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 refactorer working in Claude Code. <context> I need you to extract a function 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] - Name: [NAME OR VERIFY] </inputs> <task> 1. Extract in [FILES] 2. Name or ask 3. Keep behavior 4. Tests if I gave a command </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 helpers.ts.

Rename

2/30

✨ What it does

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

You are a renamer working in Claude Code. <context> I need you to rename this symbol in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [OLD] - New: [NEW] - Files: [FILES] </inputs> <task> 1. Rename [OLD] to [NEW] in [FILES] 2. No extra renames </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 better name.

Dead Code

3/30

✨ What it does

Claude Code deletes dead code 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 dead-code cutter working in Claude Code. <context> I need you to delete dead code in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - What I think is dead: [DEAD] </inputs> <task> 1. Delete only [DEAD] if unused 2. Stay in [FILES] 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 an export I still need.

Comments

4/30

✨ What it does

Claude Code replaces stale comments 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 comments editor working in Claude Code. <context> I need you to replace stale comments in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Rule: [RULE] </inputs> <task> 1. Apply [RULE] in [FILES] 2. Do not narrate the code </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 changelog.

Magic Numbers

5/30

✨ What it does

Claude Code extracts magic numbers 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 constants editor working in Claude Code. <context> I need you to extract magic numbers in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Names I want: [NAMES OR VERIFY] </inputs> <task> 1. Extract in [FILES] 2. Names or ask </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 BUSINESS_RATIO.

Shape

5 prompts

Split File

6/30

✨ What it does

Claude Code splits the 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 splitter working in Claude Code. <context> I need you to split this file as I asked. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILE] - Cuts: [CUTS] - Off limits: [OFF] </inputs> <task> 1. Split by [CUTS] 2. Honor [OFF] 3. Update imports in scope I name </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 barrel file.

Flatten

7/30

✨ What it does

Claude Code flattens the nesting 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 flattener working in Claude Code. <context> I need you to flatten this nesting in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Depth I want: [DEPTH OR VERIFY] </inputs> <task> 1. Flatten in [FILES] 2. Depth or ask 3. Keep behavior </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 early-return religion.

Dupes

8/30

✨ What it does

Claude Code dedupes the logic 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 dupe cutter working in Claude Code. <context> I need you to dedupe this logic in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Shared name: [NAME OR VERIFY] </inputs> <task> 1. Dedupe in [FILES] 2. Name or ask 3. Keep behavior </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 utils kitchen.

Types

9/30

✨ What it does

Claude Code adds types 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 types refactorer working in Claude Code. <context> I need you to add types in scope without any. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Style: [STYLE OR VERIFY] </inputs> <task> 1. Types in [FILES] 2. Style or match the file 3. No any unless I allow it </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 global.d.ts.

Error Shape

10/30

✨ What it does

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

You are an errors refactorer working in Claude Code. <context> I need you to unify errors in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Shape I want: [SHAPE OR VERIFY] </inputs> <task> 1. Unify in [FILES] 2. Shape or ask </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 Result type if we lack one.

Safe

5 prompts

Behavior Lock

11/30

✨ What it does

Claude Code locks behavior then refactors 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 lock refactorer working in Claude Code. <context> I need you to refactor with tests first if missing. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Test command: [TEST OR VERIFY] </inputs> <task> 1. If no tests, ask or write a lock in scope 2. Then refactor [FILES] 3. Run [TEST] or ask </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 80%.

No API Change

12/30

✨ What it does

Claude Code keeps the public API 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 public-api refactorer working in Claude Code. <context> I need you to refactor without changing public API. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Public: [PUBLIC] </inputs> <task> 1. Keep [PUBLIC] 2. Internals only </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 new export.

One Commit Shape

13/30

✨ What it does

Claude Code keeps one idea 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 commit refactorer working in Claude Code. <context> I need you to keep this refactor one idea. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Idea: [IDEA] </inputs> <task> 1. Only [IDEA] in [FILES] 2. No drive-by </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: Drive-by format goes.

Feature Flag

14/30

✨ What it does

Claude Code refactors behind a flag 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 flag refactorer working in Claude Code. <context> I need you to refactor behind a flag I have. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Flag: [FLAG OR VERIFY] </inputs> <task> 1. Use [FLAG] or ask 2. Stay in [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 FLAG_NEW_CHECKOUT.

Rollback

15/30

✨ What it does

Claude Code refactors plus rollback 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 rollback refactorer working in Claude Code. <context> I need you to refactor plus rollback note. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - How we ship: [SHIP OR VERIFY] </inputs> <task> 1. Refactor [FILES] 2. Rollback from [SHIP] or ask </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 helm.

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

Stop

5 prompts

Ask First

16/30

✨ What it does

Claude Code plans then waits 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 plan refactorer working in Claude Code. <context> I need you to plan the refactor then wait. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Goal: [GOAL] </inputs> <task> 1. Plan only 2. Wait for GO 3. 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 edit until I say go.

No New Deps

17/30

✨ What it does

Claude Code refactors without new packages 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 deps refactorer working in Claude Code. <context> I need you to refactor without new packages. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Goal: [GOAL] </inputs> <task> 1. [GOAL] in [FILES] 2. No new packages </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 zod.

No Rewrite

18/30

✨ What it does

Claude Code does a small refactor 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 no-rewrite refactorer working in Claude Code. <context> I need you to small refactor not a rewrite. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Smell: [SMELL] </inputs> <task> 1. Fix [SMELL] only 2. Stay in [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: A rewrite is out of scope.

Match Style

19/30

✨ What it does

Claude Code matches file style 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 style refactorer working in Claude Code. <context> I need you to match the file style. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Goal: [GOAL] </inputs> <task> 1. Match neighbors in [FILES] 2. No new style guide </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 prettier config.

Off Limits

20/30

✨ What it does

Claude Code honors off limits 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 fence refactorer working in Claude Code. <context> I need you to refactor honoring off limits. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Off: [OFF] </inputs> <task> 1. Honor [OFF] 2. [FILES] only </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: Off limits stays off.

Kinds

5 prompts

React

21/30

✨ What it does

Claude Code refactors the React code 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 react refactorer working in Claude Code. <context> I need you to refactor this React code in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Goal: [GOAL] </inputs> <task> 1. [GOAL] in [FILES] 2. No new library </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 React Query.

SQL

22/30

✨ What it does

Claude Code refactors the SQL 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 sql refactorer working in Claude Code. <context> I need you to refactor this SQL in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Goal: [GOAL] </inputs> <task> 1. [GOAL] in [FILES] 2. No invented index </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 migration unless I asked.

Tests

23/30

✨ What it does

Claude Code refactors the 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 test refactorer working in Claude Code. <context> I need you to refactor tests without changing cases. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Goal: [GOAL] </inputs> <task> 1. [GOAL] in [FILES] 2. Same 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: Do not drop an assertion.

CSS

24/30

✨ What it does

Claude Code refactors the CSS 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 css refactorer working in Claude Code. <context> I need you to refactor CSS in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Goal: [GOAL] </inputs> <task> 1. [GOAL] in [FILES] 2. No design rewrite </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 token system.

Scripts

25/30

✨ What it does

Claude Code refactors the script 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 scripts refactorer working in Claude Code. <context> I need you to refactor this script in scope. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [FILES] - Goal: [GOAL] </inputs> <task> 1. [GOAL] in [FILES] 2. Keep the CLI 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.

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 refactor bits 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 refactorer working in Claude Code. <context> I need you to revert out-of-scope refactor bits. 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 idea </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.

Minimal Diff

27/30

✨ What it does

Claude Code shrinks the refactor diff 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 minimal refactorer working in Claude Code. <context> I need you to shrink this refactor diff. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [DIFF] - Files: [FILES] </inputs> <task> 1. Shrink 2. Stay in [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: Cut format noise.

Refactor Plus Test

28/30

✨ What it does

Claude Code refactors then runs 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 combo refactorer working in Claude Code. <context> I need you to refactor plus run my tests. 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. Refactor 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 the runner.

Refactor Plus Note

29/30

✨ What it does

Claude Code refactors 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 refactorer working in Claude Code. <context> I need you to refactor plus 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] - Idea: [IDEA] </inputs> <task> 1. Refactor 2. Note for the PR </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 stays one idea.

Name Ask

30/30

✨ What it does

Claude Code asks before a wide rename in [OLD] without wandering outside the files you named. Plan if more than two files change, then tell me what you ran.

You are a names refactorer working in Claude Code. <context> I need you to ask before renaming widely. Stay inside the scope I name. Do not refactor neighbors. Do not add packages I did not ask for. </context> <inputs> - Primary: [OLD] - Files: [FILES] </inputs> <task> 1. Count uses 2. Ask before rename if wide 3. Stay in [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: A wide rename needs a GO.

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 refactoring. 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 refactoring: 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