.cursor — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited .cursor (MCP Server) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
AI helps you build it. AMB helps you launch it.
The production-ready foundation for shipping mobile and web apps. Auth, deployment, monitoring, and CI/CD — already wired together.
📱 Mobile app — iOS and Android from one codebase (Expo + Tamagui)
🌐 Web app — Next.js + Tailwind CSS
⚡ Backend — Supabase: auth, database, storage, and edge functions
⏱️ Background jobs — Trigger.dev for async work
📊 Monitoring — Sentry, PostHog, and Langfuse wired in from day one
🤖 Agent-native — Built-in AI context files under .agents/skills/ so Cursor, Claude, and Copilot understand the framework out of the box
🚀 Ship from `main` — Merge to main runs path-filtered deploys (Vercel, Supabase, Trigger.dev, mobile) when those parts of the repo change
For mobile development (macOS):
Option A: Click "Use this template" at the top of the repo to create your own copy with a clean git history.
Option B: Clone directly:
git clone https://github.com/ampli-group/agentic-mobile-blueprint.git my-app
cd my-app
rm -rf .git && git init# 1. Install mise (if not already installed)
curl https://mise.run | sh
# 2. Restart your terminal or run:
eval "$(mise activate zsh)" # or bash
# 3. Trust this project and install tools (Node, Deno, Supabase CLI)
mise trust
mise install
# 4. Install npm dependencies
mise run install
# 5. Start Supabase
mise run supabase
# 6. Configure environment variables
mise run configureOpen separate terminals:
# Terminal 1: Supabase Edge Functions
mise run backend
# Terminal 2: Web app
mise run frontend
# Terminal 3: Mobile app
mise run mobile
# › Press a │ open Android
# › Press i │ open iOS simulator
# › Press w │ open web
# Terminal 4: Background jobs (optional)
mise run triggerAfter creating your project, update these files:
mobile/app.json — change name, slug, and schememobile/build.json — update if using EAS Build.github/workflows/ci.yml — update badge URLs if your repo name differsmise run configure to regenerate local env files after Supabase or network changesmise run setup:prod when you are ready for production credentials and servicesRun the production setup wizard to configure services and env files:
mise run setup:prodThis interactive script guides you through Supabase, Vercel, Expo, optional Trigger.dev/Sentry, and more. Expect about 15–30 minutes. After the wizard, use [dev-to-prod.md](./dev-to-prod.md) for manual first-pass deployment steps.
Merging to `main` deploys what changed (each workflow is path-filtered).
git push origin main
# ✅ Frontend → Vercel (if frontend/ changed)
# ✅ Supabase → DB + functions (if supabase/ changed)
# ✅ Trigger.dev → jobs (if trigger/ changed)
# ✅ Mobile → iOS + Android (if mobile/ changed)main when matching paths change[dev-to-prod.md](./dev-to-prod.md) is the full production deployment guide (services, secrets, troubleshooting).
| Layer | Technology | Purpose |
|---|---|---|
| Mobile | React Native + Expo + Tamagui | iOS and Android from one codebase |
| Web | Next.js + Tailwind CSS | Server-rendered web app |
| Backend | Supabase | Auth, Postgres database, storage, edge functions |
| Jobs | Trigger.dev | Background and async work |
| Monitoring | Sentry, PostHog, Langfuse | Errors, analytics, LLM observability |
| AI | OpenAI, Anthropic, Gemini | Multi-provider LLM support via Langfuse |
Each tool has one job. Supabase handles the entire backend. Expo gives you native mobile from one codebase. Next.js powers the web. Trigger.dev runs background work. Observability is wired in from the start so you do not retrofit it later. See spec.md for the full rationale.
This framework is built for AI-assisted development. The AGENTS.md file gives coding agents full context on the codebase structure, conventions, and verification workflows. The .agents/skills/ directory contains pattern guides for each layer:
Your AI coding assistant (Cursor, Claude, Copilot) can use these to produce code that follows the framework's conventions without you having to explain the architecture.
├── .mise.toml # Tasks and tool versions
├── AGENTS.md # AI agent context
├── .agents/skills/ # AI coding pattern guides
├── frontend/ # Next.js + Tailwind CSS
├── mobile/ # React Native + Expo + Tamagui
├── supabase/ # Config, migrations, edge functions
├── trigger/ # Background jobs
├── scripts/ # Setup and configuration scripts
└── docs/ # Tutorials and guides| Doc | Purpose |
|---|---|
| Available Commands | All mise commands and aliases |
| Environment Variables | Env files, naming conventions, and ports |
| dev-to-prod.md | Production deployment guide |
| spec.md | Architecture rationale |
| CONTRIBUTING.md | How to contribute |
| SECURITY.md | Security policy |
We welcome contributions — whether it is a bug fix, a new skill, or a documentation improvement. Check out CONTRIBUTING.md to get started, or browse issues labelled good first issue if you are looking for a place to jump in.
For security vulnerabilities, see SECURITY.md.
MIT — see LICENSE.
Built and maintained by Ampli Group — we ship our own production apps on this framework every day.
If this saved you setup time, a ⭐ helps others find it.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.