Route discovery

Frameworks matter for only one step: finding the list of routes. skillgendev tries these sources and merges what they find into .skillgendev/routes.json.

Source	Covered today
Folder routing (built in, from folder hint packs)	Next.js (app and pages router), Nuxt, SvelteKit, Astro
Routes declared in code (ast-grep router hint packs)	React Router, route objects (Vue Router, Angular, data routers), Express, Flask, Django, Laravel, Rails, Spring MVC, ASP.NET Core attributes, Phoenix
Graphify's own route nodes	ASP.NET Razor Pages (@page "/x"), Dart / Flutter navigation
Claude Code reading the code graph	Any other stack: Go, Rails resources, conventional MVC routes, custom routers, CMSs
You	skillgendev routes --route /leads/import

Tested end to end on React Router, Next.js and Flask apps. The Rails, Spring, ASP.NET, Phoenix and Laravel hint packs are tested on small sample apps, and the Claude step was tried live on Rails and Go samples.

How the Claude step works

Graphify does not name routes in most stacks, but its links show hubs: files such as a router that link to many controllers, views or pages. skillgendev ranks those likely router files and gives Claude Code the map ("this file links to these files"). Claude reads them read-only and returns the routes with their entry files.

It runs automatically when no other source finds a route, or on demand:

skillgendev routes --all --ask-claude    # also let Claude list routes
skillgendev routes --all --no-claude     # never call Claude

The prompt is saved in .skillgendev/route-listing-prompt.md so you can see what was asked.

Merging into routes.json

skillgendev routes --all
skillgendev routes --route /leads/import --route /audiences/:id   # add by hand, repeatable
skillgendev routes --all --include "campaigns/*" --exclude "api/*"

Each entry looks like this:

{
  "route": "/campaigns/:campaignId/report",
  "skill": "acme-campaigns-report",
  "title": "",
  "entry_files": ["app/campaigns/[campaignId]/report/page.tsx"],
  "sources": ["folder:nextjs-app"],
  "has_form": null,
  "skip": false,
  "name_locked": false
}

You may edit skill, title, skip and open. Running routes --all again updates the file from the current code:

• A route's code files are replaced by the ones found now.
• Routes the code no longer has are removed, unless you named, skipped or generated them.
• Names you chose are locked and never change; clashing generated names get -2, -3.
• has_form is filled in by generate: false routes are skipped later unless you pass --force.

Routes are normalised to one form, /a/:id/b, whatever the source wrote ([id], <id>, {id}). Dependency and build folders (node_modules, dist, hidden folders) are never searched. Apps inside one repository, such as an admin app and a main site side by side, are all found.

Concrete pages

A route you give as one concrete page (/audiences/7) is linked to the pattern found in the code (/audiences/:id). The pattern with the most fixed segments wins, and a tie links nothing.

Teaching skillgendev a router

Add a small YAML file under hintpacks/; no Python needed. See Hint packs and Adding framework support.