Scan the repository and build the architecture model (routes, entities, tasks, module dependency graph, findings). Run once at session start and after structural changes.

Compact architecture summary: modules, route prefixes, domain entities, hotspots, visibility score. Use this as ground truth for where things live before searching the codebase.

Given a file or top-level module path, report what depends on it (fan-in), what it depends on (fan-out), and the routes/entities it contains. Call BEFORE editing anything load-bearing.

Deterministic risk findings: hardcoded secrets, raw SQL bypassing the ORM, auth tokens in localStorage, routes without auth guards.

Score the repo's production readiness 0-100 across 8 deterministic dimensions (auth coverage, secrets hygiene, client token storage, data-layer discipline, test footprint, migration discipline, ops readiness, docs/API contract) with per-dimension evidence and the top fixes ranked by points lost. Same repo always scores the same.

Generate architecture diagrams as Mermaid sources: module dependency graph, domain-model ER diagram, and HTTP route map. Also writes .archiet/diagrams/ (3 .mmd files + a self-contained HTML viewer). Paste the Mermaid into any README/wiki — GitHub, GitLab, Notion and VS Code render it natively.