Loading…
Loading…
This is exactly what Archiet produced from the description below. No manual tweaking, no prompt engineering. Flask backend, Next.js + shadcn frontend, Expo mobile app, seven ADRs, GDPR DPIA populated from the entity model, Mermaid C4 diagrams, traceability matrix, cost estimate, security posture. Quality score: 100/100.
"A multi-tenant task management SaaS. Users belong to workspaces, each workspace has projects, each project has tasks. Tasks have a title, description, due date, status (todo / in progress / done), priority, and assigned user. Activity feed per project. Email notifications on task assignment. Integration with Slack for notifications. GDPR flag. Target: small teams, up to 50 users per workspace."
docs/ treeThe architect-facing handoff pack that ships next to the code.
docs/
├── README.md # 5-question handoff index
├── decisions/
│ ├── ADR-0001-backend-stack-flask-nextjs.md
│ ├── ADR-0002-frontend-stack-shadcn.md
│ ├── ADR-0003-database-postgresql.md
│ ├── ADR-0004-auth-jwt-httponly-cookies.md
│ ├── ADR-0005-orm-sqlalchemy-alembic.md
│ ├── ADR-0006-integration-slack.md
│ ├── ADR-0007-compliance-gdpr.md
│ └── INDEX.md
├── diagrams/
│ ├── context.mmd # C4 system context
│ ├── container.mmd # C4 containers
│ ├── data-model.mmd # ERD
│ ├── request-flow-task-list.mmd # sequence per screen
│ ├── state-task.mmd # state machine
│ └── INDEX.md
├── traceability.md # story → FR → file chain
├── traceability.csv # audit export
├── compliance/
│ ├── dpia.md # GDPR Article 35
│ └── INDEX.md
├── cost-estimate.md # monthly TCO + growth
├── security-posture.md # attack surface + checklist
└── performance-notes.md # scaling + observabilityEvery ADR is a real MADR-format document, every .mmd renders in GitHub natively, compliance/dpia.md enumerates the actual PII fields from the entity model. Browse a live sample.
app/
├── main.py
├── models/
│ ├── user.py
│ ├── workspace.py
│ ├── project.py
│ ├── task.py
│ └── activity.py
├── blueprints/
│ ├── auth_bp.py
│ ├── workspace_bp.py
│ ├── project_bp.py
│ └── task_bp.py
├── services/
│ └── slack_notifier.py
└── schemas/ ...
alembic/versions/
├── 0001_initial.py
├── 0002_add_task_state_machine.py
└── 0003_add_activity_feed.py
frontend/app/
├── (protected)/
│ ├── dashboard/page.tsx
│ ├── projects/page.tsx
│ ├── projects/[id]/page.tsx
│ ├── tasks/page.tsx
│ └── settings/page.tsx
├── login/page.tsx
└── register/page.tsx
mobile/
├── app/(tabs)/
│ ├── index.tsx
│ ├── tasks.tsx
│ └── settings.tsx
└── eas.jsonAlembic migrations for every model change. State-machine transitions per stateful entity. Next.js App Router with shadcn components. Expo tabs with the same auth model. Slack integration wired via the vendor catalog — credentials stay encrypted inintegration_credentials.
docs/compliance/dpia.md lists every PII-candidate field (user email, task assignee, activity feed user) with detection source. Lawful-basis and retention-period fields are explicitly marked TBD — human judgement stays with your DPO, not hallucinated by the generator.
docs/diagrams/state-task.mmd renders the todo → in-progress → done transitions as astateDiagram-v2 in GitHub. Alembic migration 0002_add_task_state_machine.py enforces the valid transitions at the database layer.
Every user story maps to at least one functional requirement; every requirement maps to a screen and an entity; every entity maps to generated model / route / schema files. Orphan-FR detection is on. The CSV export feeds straight into audit tooling.
Monthly baseline on managed hosting: $155 (compute + Postgres + Redis + Slack SDK + GDPR overhead). Scaling projections: $385/mo at 10k MAU, $1,200/mo at 100k. When to add read replicas, when to extract bounded contexts — all in docs/cost-estimate.md.
Free trial. No credit card. Describe your product, pick your stack, hit generate. The handoff pack lands in your inbox as a ZIP you can open in your editor.