DEVELOPER ONBOARDING
Ship in days, not months, on an unfamiliar codebase.
Mapnostics auto-generates an onboarding guide from your code and its history: the architecture, the entry points, a ranked learning path with time estimates, change recipes for common tasks, and troubleshooting playbooks — all cited back to real files.
The repository implements a Next.js commerce platform on a Supabase backend, serving checkout, admin and analytics from one App Router application.
Server components render the customer-facing pages, client components own the checkout stepper and the admin dialogs, and middleware verifies the session before any protected route runs. Tailwind CSS and a small token layer keep the two halves consistent.
Startup flow: No startup hook is declared, so there is no boot sequence to read from the code.
Background jobs: Nothing is scheduled on a cron entry or a queue in this repository.
Event flow: No event bus or emitter is declared. Inferring one from function names would be a guess presented as a map.
Shutdown: Serverless functions are terminated rather than shut down, so there is no hook to find.
The application is entered here. Pages and API routes are resolved by path, so nothing in the repository imports them.
+ 16 more files — these are just the most depended-on to read first
Configuration and the request gate. Middleware protects every route the matcher names.
+ 2 more files — these are just the most depended-on to read first
Support code centred on supabase/server.ts, which 21 other files import.
+ 12 more files — these are just the most depended-on to read first
The client surface: the checkout stepper, the admin dialogs, and the shared data table.
+ 8 more files — these are just the most depended-on to read first
Verifies the Supabase session on every protected route and redirects anonymous visitors to the login page.
Renders the four-step checkout and holds the cart for the session.
Creates an order, charges it through Stripe, and writes the orders and order_items tables.
Lists customers with their orders, and opens the detail dialog for one of them.
Unused exports and unreachable branches across 12 files.
Reachable code nothing has called in 90 days of history.
A credential-shaped string committed to the repository.
Not yet scanned (unknown, not clean): unused asset, orphan table.
Drives the four checkout steps and validates each one before it advances.
Renders the checkout stepper and holds the cart for the session.
Sorts, filters and paginates every table in the admin area.
Shows one customer with their orders and invoices.
Add a SQL migration for the column — the schema here is defined by migrations, not by the code.
Update the code that writes orders, so the new column is set where rows are inserted or updated.
Surface it where orders are read, so the new column reaches whatever consumes the row.
No feature flags were found, so there is no flag file to update in this repository.
No webhook receiver was found, so this repository has no webhook surface to extend.
No scheduled entry point (cron) was found, so this repository has no background job recipe.
A change broke something that seemed unrelated
Start with the most-imported files. A change to one ripples to every file that imports it. These have the widest blast radius, so check their importers before and after your change.
A request 404s or never reaches your handler
Middleware runs first and can redirect or block a request before any handler sees it. Routes resolve by file path, not by import — nothing in the code calls them.
A query returns nothing, or a write silently does not persist
Check the files that own each table. Row-level security and a mismatched filter both show up as empty results, not as errors.
This guide is built entirely from your code and its history. Some things a new engineer needs are not in the code at all — they are decisions people made — so Mapnostics does not guess at them rather than present a confident guess as fact:
A plain-English explanation of what every file does.
The dependency graph, architecture layers and schema.
See the blast radius before you change a file.
Unused code that is safe to remove.
Written from your code
Every section is built from the repository and its git history, and badged verified, estimated or inference so a new engineer knows what to trust.
A path, not a file dump
The learning path is ranked by how much the rest of the codebase depends on each file, with an hour estimate on every step.
Honest about its limits
Business context, on-call paths and team ownership live in people's heads, not the code — the guide says so instead of guessing.
Generate an onboarding guide in under 2 minutes.
Analyze my codebase →No credit card required · Free tier available