R RomantiCode
Install Free
For AI-generated, vibe-coded & legacy projects

Understand and prepare AI-generated code before cleanup, review, or refactor.

Your app works. Before you clean up, refactor, or hand it off — you need to understand the architecture, document what exists, and prepare an AI-ready context pack.

Install LegacyDoc AI Generate an AI Code Audit Report
audit-report.md · taskflow-mvp · sample

Project overview

taskflow-mvp · Next.js 15 + Supabase · 12k LOC · built with Cursor + Claude Code. Working MVP, no tests, no docs.

Architecture map

app/
├── (auth)/        # sign in / sign up
├── dashboard/
│   ├── tasks/     # CRUD + filters  ← largest surface
│   ├── projects/  # grouping
│   └── settings/  # preferences
├── api/
│   ├── tasks/route.ts
│   └── webhooks/
└── lib/
    ├── supabase.ts  # mixed server/client usage
    ├── auth.ts
    └── ai.ts        # behind feature flag

Areas to inspect

  • app/api/tasks/route.ts — input parsing
  • lib/supabase.ts — server/client boundary
  • TaskList.tsx — 380 lines, split candidate
  • lib/ai.ts — feature flag + timeout

Cleanup priorities

  1. Split TaskList into smaller components
  2. Move DB-only code to server modules
  3. Unify API input parsing
  4. Document required env vars
Runs inside VS Code BYOK to your AI provider No code storage by RomantiCode Audit-ready context pack

Your app works. The codebase is the question.

Six things every AI-generated or vibe-coded project tends to share — and why cleanup without context becomes guesswork.

Missing docs — no one knows what each module does
No architecture map — hard to see how pieces connect
Duplicated logic — AI tools often repeat patterns
AI coding tools lack project context
Cleanup has no plan — where do you even start?
Handoff is painful — specialists need context you don't have

Products and tools

A small but focused set of resources for developers working with AI-generated, vibe-coded, and legacy codebases.

Product

LegacyDoc AI

VS Code extension that generates JSDoc, Markdown docs, Mermaid architecture diagrams, and AI-ready context packs from your local codebase.

Learn more

Tool

AI Code Audit Report

Generate an audit-ready context pack for AI-generated code before cleanup, refactor, review, or launch.

Learn more

Example

Sample audit report

See what a finished AI code audit report looks like — project overview, architecture map, areas to inspect, cleanup priorities.

Learn more

Use case

Vibe Code Cleanup Readiness

Prepare your vibe-coded app for cleanup with architecture maps, module summaries, and cleanup priorities.

Learn more

Resource

AI Code Audit Checklist

A practical checklist to review architecture, dependencies, docs, tests, and cleanup priorities before shipping AI-generated code.

Learn more

How it works

A simple loop you can run before cleanup, refactor, or handoff — locally inside VS Code, with your own API key.

  1. Step 1

    Open your codebase locally in VS Code

  2. Step 2

    Generate docs, context pack, and architecture map

  3. Step 3

    Review cleanup notes and priorities

  4. Step 4

    Share context with Claude Code, Cursor, or Codex

  5. Step 5

    Refactor in small, reviewable steps

What this is — and what it's not

Honest boundaries help you (and your team) trust what you're building on.

What it is

  • — Generates JSDoc, Markdown docs, and Mermaid architecture diagrams
  • — Produces an audit-ready context pack for your codebase
  • — Runs locally in VS Code, BYOK to your AI provider
  • — Designed to be shared with Claude Code, Cursor, or Codex

What it's not

  • — It does not replace a professional security audit
  • — It does not automatically fix every issue in your code
  • — It does not guarantee production readiness
  • — It does not store or proxy your code on RomantiCode servers

Ready to understand your codebase?

Free to install. Runs inside VS Code. Works with your own API key.

Install LegacyDoc AI Free Generate an AI Code Audit Report