Codebase to Full PRD
Turns a whole codebase into an as-built PRD: reconciled inventory of pages, endpoints, jobs, roles, file-traced rules, and coverage proof.
Before & after
Same prompt, same agent — the only difference is the skill.
You could analyze the code to extract what the application does. Start by looking at the routes to understand the main pages and flows. Then examine the components to see what UI elements exist and how they're structured. Check the state management to understand what data the application tracks. Look at the API calls to see what integrations and endpoints are used. Then write up a document that describes all of these things so that someone reading it would understand how the application works. This might take a while depending on the size of the codebase, and you'd probably want to focus on the most important parts first.
I'll analyze your React codebase and generate a complete PRD covering architecture, data model, user flows, and API integration. DELIVERABLE Single PRD document structured for engineers and product teams to rebuild or maintain the app without examining code. EXTRACTION COVERS ROUTES & PAGES: Public/authenticated paths, route parameters, redirect logic, named page layouts COMPONENTS: Visual hierarchy, UI element inventory, stateful vs. presentational split, prop contracts STATE MANAGEMENT: Global store schema, local component state, state mutations, data flow patterns FORMS & VALIDATION: Input fields by page, validation rules, error handling, submission flows API ENDPOINTS: Service URL patterns, HTTP methods, request/response schemas, error codes, auth requirements EXTERNAL INTEGRATIONS: Third-party services (Stripe, Auth0, tracking, etc.), handshake flow, failure modes OUTPUT STRUCTURE Executives summary (one page: core functionality and user personas) -> Use cases and flows (customer journeys with decisions) -> Page-by-page spec (every public/private route listed with field-level detail) -> Data model (ERD-style description of entities, relationships, cardinality) -> API reference (endpoint signatures) -> Feature flags and edge cases. This captures enough detail that a new engineer can implement missing features or maintain the codebase without asking constant questions.
About this skill
Find out everything an undocumented application actually does before you rewrite, migrate, or take it over. You get an as-built PRD where every requirement traces to a source file and a coverage appendix shows nothing was skipped.
What it does
- Detects the stack and finds where routes, endpoints, schema, jobs, and access rules are declared across common frameworks
- Builds and reconciles an inventory of pages, endpoints, background jobs, entities, and integrations, flagging dead or hidden surfaces
- Documents each module: user flows, fields and validation, business rules, and state transitions
- Produces a roles and permissions matrix from server-side guards, noting where the UI hides what the API still allows
- Lists non-functional limits, notifications, and discrepancies labeled bug, undocumented policy, or product decision
What you get
A multi-section PRD (or one file per module for large repos), full inventory tables, and a coverage appendix of what was read, skimmed, and not read.
Best for
Teams inheriting or acquiring a codebase, rewrite and migration leads, and technical due diligence.
SKILL.md preview
---
name: code-to-prd
… (sign up to view the full skill)More business & ops skills
View all Business & Ops skills →Product Practice Diagnostic
Trace a product-team symptom (unused features, roadmap churn, priority fights) to the broken stage of the loop and fix one thing first.
Feature Launch Go-to-Market Plan
Plan one feature launch: tier, plan access, audience messages, readiness gates for support, sales, billing, and docs, and success metrics.
Pricing Tier and Packaging Diagnosis
Finds why your pricing tiers underperform and redesigns value metric, fences, price gaps, and upgrade triggers with a migration plan.
Product Specs, Roadmaps, and Research Decisions
Turn customer evidence into testable PRDs, dependency-aware roadmaps, and stakeholder updates with clear decisions and tradeoffs.