Graph Shapes by App Type

A crawl maps your app into a Knowledge Graph. Its shape reflects how the app gates access.

No auth — marketing site

One public zone, fully connected from the entry point.

(/) ─→ /about
  ├─→ /pricing ─→ /signup
  ├─→ /blog ─→ /blog/post
  └─→ /contact

Public + auth

A public shell and a gated app, joined at the login page. The login → app edge is only crossed after authentication.

PUBLIC                          AUTHED
(/) ─→ /about
  ├─→ /pricing
  └─→ /login ──[ auth ]──→ /dashboard ─→ /tasks
                                 └─→ /settings

Auth only

The entry point redirects to login, so the graph depends on credentials:

No / bad creds:   (/) ─→ /login ⊣        AUTH failure — login is the only node
Valid creds:      (/) ─→ /login ─auth→ /dashboard ─→ /tasks ─→ /reports

An auth wall reached without valid credentials is reported as an AUTH failure — not an empty success.

Auth + tiered permissions

Roadmap

Multi-role crawling is planned. Today a crawl runs as a single credential set.

One page set, role-gated. The graph is the union of what every role can reach, with each node tagged by the roles that can see it (denials recorded too).

/dashboard       {admin, member, viewer}
  ├─ /tasks      {admin, member, viewer}
  ├─ /reports    {admin, member}      ← viewer denied
  └─ /users      {admin}              ← member, viewer denied

Built by crawling once per credential tier, then merging the results.

What the graph records

On	Attribute	Meaning
Node	zone	public or authed
Node	status	200, redirect-to-login, or denied
Node	accessible by	roles that can reach it (tiered apps)
Edge	crosses auth	the login → app transition
Run	outcome	auth-only with no credentials → AUTH failure

Related

• Crawling your application — auth config, exclude patterns, depth limits
• Knowledge Graph overview