JPeetz/agent-skills / skills/development/app-scaffolding
app-scaffolding
VerifiedUse this skill when the user has decided to build a mobile app and needs a complete production blueprint — not brainstorming, not feature ideas, but an actual build brief. Covers both iOS and Android with exact tech stack, screen specs, monetization, and a 30-day sprint plan. Use when the user says things like "Scaffold this app" "Give me a build plan" "Write the PRD" "How do I build this?" "Create a technical spec" "Blueprint this idea" even if they don't mention "scaffolding" or the number of sections.
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 372 B |
| SKILL.md | skill md | 5.0 KB |
| evals/evals.json | config | 2.6 KB |
| references/corrections.md | markdown | 1.6 KB |
evals/evals.json
View on GitHub{
"skill_name": "app-scaffolding",
"evals": [
{
"id": 1,
"prompt": "BiteSaver — recipe saver app with AI meal planning. It won the app discovery. $4.99/mo, social cooking challenges, iOS + Android. Give me the full build blueprint.",
"expected_output": "12-section scaffold including: ASO-optimized name, SwiftUI screen specs, Jetpack Compose screen specs, RevenueCat paywall logic, DB schema (SQL), 30-day sprint plan, launch checklist, anti-patterns. Must include exact hex colors, typography specs, and at least 6 key animations. Must specify exact tools: fal.ai FLUX for images, video_generate for video, music_generate for audio. Must not reference Runway or Luma as primary tools.",
"should_trigger": true
},
{
"id": 2,
"prompt": "I need a PRD for a new mobile app called FitTracker",
"expected_output": "Skill should activate — 'PRD' and 'mobile app' match. Must produce at minimum: tech stack (SwiftUI/Jetpack Compose), revenue model, 30-day sprint. If output is only a generic one-page PRD without tech specs, test fails.",
"should_trigger": true
},
{
"id": 3,
"prompt": "Write me a script that scrapes product data from Amazon",
"expected_output": "Skill must NOT activate. This is a web scraping task, not app scaffolding. If activated, test fails. Near-miss: involves 'build' but not mobile app building.",
"should_trigger": false
},
{
"id": 4,
"prompt": "I want to build a Chrome extension that summarizes emails. Blueprint this.",
"expected_output": "Skill must NOT activate. Chrome extension is not a mobile app. Agent should handle without scaffolding. If activated, test fails. Near-miss: uses 'blueprint' but wrong platform.",
"should_trigger": false
},
{
"id": 5,
"prompt": "What are the first steps to building a meditation timer app? Just high-level, no need for a full spec.",
"expected_output": "Should trigger — user is asking to build a mobile app. Even with 'no need for full spec,' scaffolding skill should adapt output depth. Must produce at minimum: tech stack recommendation, monetization model, and launch plan. Quality gate: if output is over-scaffolded for a 'just high-level' request, partial pass — skill should match depth to request.",
"should_trigger": true
},
{
"id": 6,
"prompt": "How do I write a technical specification for my SaaS product?",
"expected_output": "Skill must NOT activate. 'Technical specification' + 'SaaS' means this is about a web SaaS product, not a mobile app. If activated, test fails — scope creep.",
"should_trigger": false
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |