Home › Rules › Fixing a messy vibe-coded project
Fixing a messy vibe-coded project rules: AGENTS.md, CLAUDE.md & Cursor rules
When a project is already messy, the AI's instinct is to rewrite it. That's how you lose the parts that worked. These rules force read, test, fix one thing, verify.
Stack these rules assume: Any existing project that "mostly works". Different stack? Customize it in the generator.
Want the full step-by-step kit?
Codebase Fixer Agent ($8): A strict audit, safety net, fix one thing, verify process for your agent, plus a bug report template and refactor recipes. Stops the "it rewrote half my app" problem.
Get the kit for $8 on Gumroad → See what's inside
or get all 5 kits for $25 (save $14) →
One markdown file, instant download. 7-day refund, no questions asked.
Quick start
- Copy the
AGENTS.mdbelow into your project's root folder. - Fill in the three
[bracketed]lines under "Project". Delete any rule that doesn't fit. - Start your first chat with: "Read AGENTS.md and follow it for this whole session."
AGENTS.md (works across tools)
Read by OpenAI Codex, Cursor, GitHub Copilot, Claude Code (when there's no CLAUDE.md) and many other agents.
# AGENTS.md: Fixing a messy vibe-coded project
Instructions for AI coding agents working on this project. Replace anything in [brackets].
## Project
- What it is: [one sentence: what you are building and who it is for]
- Stack: Any existing project that "mostly works"
- Version 1 only needs to: [the ONE thing it must do]
## Commands
- First, prove it runs: Run the app exactly as the README (or package.json scripts) says, and note what works
- Commit before every fix: `git add -A && git commit -m "checkpoint"`
## Project rules
- Start by reading and summarizing the project (stack, entry points, how to run it) before changing anything.
- No rewrites. Fix one bug or one problem per task, with the smallest change that works.
- Before fixing a bug, describe how to reproduce it. After fixing, show that the reproduction now passes.
- Add a small test or a manual check list for anything you fix, so it stays fixed.
- Don't upgrade dependencies or frameworks as part of a bug fix. Suggest upgrades separately.
- Keep a short CHANGES.md log of what you changed and why.
## How to work
- Do one task per request. If my request is really several tasks, list them and do only the first.
- Before writing code, give a 3-line plan: what you'll change, which files, and how I'll test it.
- Make the smallest change that works. Don't refactor, rename, or reformat code I didn't ask about.
- Ask before adding a dependency, deleting or moving files, changing the data/database structure, or touching more than 3 files.
- If you're not sure a library, API, or setting exists in the version we use, say so and check the installed package or the docs instead of guessing.
- Never say something works unless you ran it or I confirmed it. Report errors exactly.
- Secrets (API keys, tokens, passwords) live only in environment variables or a .env file listed in .gitignore. Never print, hardcode, or commit them.
- Never run destructive commands (rm -rf, git reset --hard, force-push, dropping tables) without asking first.
## Finish every task with
1. What changed (list the files).
2. How to run or test it in under a minute, and what I should see.
3. A suggested git commit message.
Want the full step-by-step kit?
Codebase Fixer Agent ($8): A strict audit, safety net, fix one thing, verify process for your agent, plus a bug report template and refactor recipes. Stops the "it rewrote half my app" problem.
Get the kit for $8 on Gumroad → See what's inside
or get all 5 kits for $25 (save $14) →
One markdown file, instant download. 7-day refund, no questions asked.
Why each rule exists
- Fixing a messy vibe-coded project ruleStart by reading and summarizing the project (stack, entry points, how to run it) before changing anything.Why: You can't fix what you haven't understood, and neither can the AI.
- Fixing a messy vibe-coded project ruleNo rewrites. Fix one bug or one problem per task, with the smallest change that works.Why: Rewrites throw away the parts that worked and bring back bugs you already fixed.
- Fixing a messy vibe-coded project ruleBefore fixing a bug, describe how to reproduce it. After fixing, show that the reproduction now passes.Why: "Fixed" means you watched it fail and then watched it work.
- Fixing a messy vibe-coded project ruleAdd a small test or a manual check list for anything you fix, so it stays fixed.Why: Vibe-coded projects regress constantly without a safety net.
- Fixing a messy vibe-coded project ruleDon't upgrade dependencies or frameworks as part of a bug fix. Suggest upgrades separately.Why: Upgrades change lots of behavior at once, which is the opposite of a careful fix.
- Fixing a messy vibe-coded project ruleKeep a short CHANGES.md log of what you changed and why.Why: Future you (and future AI sessions) will need to know why things are the way they are.
- How to workDo one task per request. If my request is really several tasks, list them and do only the first.Why: Big multi-part requests are where AI tools change 12 files and you can't tell which change broke things.
- How to workBefore writing code, give a 3-line plan: what you'll change, which files, and how I'll test it.Why: A plan you can read in 10 seconds catches the wrong approach before it costs you an hour or a pile of credits.
- How to workMake the smallest change that works. Don't refactor, rename, or reformat code I didn't ask about.Why: Drive-by "cleanups" are the #1 cause of "it was working yesterday".
- How to workAsk before adding a dependency, deleting or moving files, changing the data/database structure, or touching more than 3 files.Why: These are the changes that are hard to undo. One quick question is cheaper than a rollback.
- How to workIf you're not sure a library, API, or setting exists in the version we use, say so and check the installed package or the docs instead of guessing.Why: AI tools confidently invent functions and config options. Making it check is the cure.
- How to workNever say something works unless you ran it or I confirmed it. Report errors exactly.Why: "Should work now" isn't a test result. You want facts, not optimism.
- How to workSecrets (API keys, tokens, passwords) live only in environment variables or a .env file listed in .gitignore. Never print, hardcode, or commit them.Why: Leaked keys get found by bots within minutes of a public push. This rule is the cheapest insurance you'll ever buy.
- How to workNever run destructive commands (rm -rf, git reset --hard, force-push, dropping tables) without asking first.Why: One bad command can delete the only copy of your work.
- Finish every task withWhat changed, how to test it, and a commit message.Why: This turns every task into a checkpoint: you test it, you commit it, and you always have a last good version to go back to.
Use it as CLAUDE.md (Claude Code)
Pick one:
- Do nothing. Current Claude Code versions read
AGENTS.mdwhen the project has noCLAUDE.md. Run/memoryin a session to check it loaded. - Import it (best if you want Claude-only extras, or you're on an older Claude Code). Create
CLAUDE.mdnext toAGENTS.md:
@AGENTS.md
## Claude Code
- Use plan mode before any change that touches more than 3 files.
- Claude only? Save the same AGENTS.md text as
CLAUDE.mdinstead. Don't keep two separate copies; they drift apart.
Heads-up: if a project has both files and CLAUDE.md doesn't import AGENTS.md, Claude Code reads only CLAUDE.md by default. More in AGENTS.md vs CLAUDE.md.
Use it as Cursor rules (.cursor/rules/*.mdc)
Cursor reads AGENTS.md too, so the file above already works. If you prefer Cursor's own Project Rules, split it into two .mdc files. The .mdc extension and frontmatter are required; Cursor ignores plain .md files in .cursor/rules.
---
description: Project overview and working rules for Fixing a messy vibe-coded project
alwaysApply: true
---
# Fixing a messy vibe-coded project
Instructions for AI coding agents working on this project. Replace anything in [brackets].
## Project
- What it is: [one sentence: what you are building and who it is for]
- Stack: Any existing project that "mostly works"
- Version 1 only needs to: [the ONE thing it must do]
## Commands
- First, prove it runs: Run the app exactly as the README (or package.json scripts) says, and note what works
- Commit before every fix: `git add -A && git commit -m "checkpoint"`
## How to work
- Do one task per request. If my request is really several tasks, list them and do only the first.
- Before writing code, give a 3-line plan: what you'll change, which files, and how I'll test it.
- Make the smallest change that works. Don't refactor, rename, or reformat code I didn't ask about.
- Ask before adding a dependency, deleting or moving files, changing the data/database structure, or touching more than 3 files.
- If you're not sure a library, API, or setting exists in the version we use, say so and check the installed package or the docs instead of guessing.
- Never say something works unless you ran it or I confirmed it. Report errors exactly.
- Secrets (API keys, tokens, passwords) live only in environment variables or a .env file listed in .gitignore. Never print, hardcode, or commit them.
- Never run destructive commands (rm -rf, git reset --hard, force-push, dropping tables) without asking first.
## Finish every task with
1. What changed (list the files).
2. How to run or test it in under a minute, and what I should see.
3. A suggested git commit message.
The project rules only attach when a matching file is in the chat (globs with alwaysApply: false). Want them on all the time? Change it to alwaysApply: true.
---
description: Project rules for Fixing a messy vibe-coded project
alwaysApply: true
---
# Project rules: Fixing a messy vibe-coded project
- Start by reading and summarizing the project (stack, entry points, how to run it) before changing anything.
- No rewrites. Fix one bug or one problem per task, with the smallest change that works.
- Before fixing a bug, describe how to reproduce it. After fixing, show that the reproduction now passes.
- Add a small test or a manual check list for anything you fix, so it stays fixed.
- Don't upgrade dependencies or frameworks as part of a bug fix. Suggest upgrades separately.
- Keep a short CHANGES.md log of what you changed and why.
Use either AGENTS.md or these .mdc files, not both, so you don't send the same rules twice. See Cursor rules vs AGENTS.md.
Related
- Best AI tool to fix and refactor existing code (which tool to use, real prices)
- AGENTS.md, CLAUDE.md and Cursor rules formats explained