Spec to Runnable Repo
Turn a written spec into a new repo in any stack: numbered requirements, real generators and lockfiles, tests, and a verified boot.
Before & after
Same prompt, same agent — the only difference is the skill.
I can definitely help you build a weather dashboard! That sounds like a great project. Before I get started, I have a few questions to clarify the scope. What weather data are you planning to use — OpenWeatherMap API, a government weather service, or something else? Should it show forecasts for multiple cities or just one location? Do you want real-time updates or will static data work? Also, I'm wondering about the UI — are you thinking a minimalist design or something more feature-rich with charts and graphs? And do you need user authentication, or should it be public-facing? I can certainly scaffold out a React project with TypeScript, but knowing these details would help me tailor it to your exact needs. What's your timeline like?
WEATHER DASHBOARD - NEXT.JS + TYPESCRIPT STARTER REPO STRUCTURE - app/page.tsx (main dashboard) - app/api/weather/route.ts (OpenWeatherMap proxy) - components/WeatherCard.tsx - components/ForecastGrid.tsx - lib/weather-service.ts (API client) - types/weather.ts (full TypeScript types) TECH STACK Next.js 15 (App Router), React 19, TypeScript 5.6, Tailwind CSS, shadcn/ui components, axios for HTTP, next-env for env vars FEATURES INCLUDED Current weather display with temp, humidity, wind speed. 5-day forecast with hourly breakdown. Search by city name. Geolocation auto-detect. Dark mode toggle. Responsive mobile-first design. Type-safe API layer. SETUP STEPS 1. npm create next-app@latest weather-dash -- --typescript 2. npm install axios shadcn-ui 3. Add NEXT_PUBLIC_WEATHER_API_KEY to .env.local 4. Copy components to app/ and components/ folders 5. npm run dev on http://localhost:3000 CORE WEB VITALS LCP target: 0.8s (image optimization built-in). CLS target: 0.05 (fixed layout, no layout shifts). Initial bundle: 45KB gzipped. All static pages pre-rendered at build time.
About this skill
AI-generated repos often look complete but ship invented package versions, silent TODO stubs, and a claim that it works that nobody ran. This skill turns your spec into a repository that installs, builds, tests, and boots, and tells you honestly what was verified.
What it does
- Extracts numbered requirements and logs every assumption instead of guessing silently
- Recommends one boring, well-supported stack for the workload
- Scaffolds with official generators and adds dependencies through the package manager
- Builds the core flow end to end with meaningful tests and safe env handling
- Runs install, lint, typecheck, tests, build, migrations, and a boot check
What you get
A runnable repository with README, env example, task runner, and CI, plus a handoff with a requirement-to-code-to-test table, assumptions, verification results, and what was not built.
Best for
Developers and technical founders turning a PRD, README, or idea list into a real starting codebase in any language.
SKILL.md preview
---
name: spec-to-repo
… (sign up to view the full skill)More development skills
View all Development skills →Performance Bottleneck Profiling: Node, Python, Go
Classifies a slowdown, gives exact profiler commands, reads the output, and proves the fix with same-conditions p95/p99 numbers.
Feature Scaffolding with Working Tests
Turn a short feature spec into code, runnable tests, and integration wiring that match your existing project.
Feature Flag Rollouts and Kill Switches
Design staged flag rollouts, verify rollback paths, and audit flag debt with explicit defaults, gates, and retirement plans.
Microcontroller Firmware and RTOS Engineering
Build and debug MCU firmware with bounded ISR/RTOS timing, safe DMA ownership, and measured power budgets.