AppLaunchFlow Screenshot Capture Skill

AI agent skill for automated App Store screenshot capture — forked from ParthJadhav/ios-marketing-capture and adapted for the AppLaunchFlow pipeline.

--- ## What changed from the original This fork extends the original iOS Marketing Capture skill with: - **Flutter support** — full screenshot capture path for Flutter apps using offline mode, `simctl io` capture, and in-process locale switching - **Auto-detection over interviews** — the skill now infers screens, locales, devices, seed data, and appearance from the codebase instead of asking the user a long questionnaire - **App Store candidate evaluation** — after capture, the skill evaluates screenshots and preselects up to 10 store-ready candidates based on visual strength, feature relevance, and composability - **AppLaunchFlow MCP handoff** — curated screenshots are automatically handed off to the [AppLaunchFlow MCP](https://github.com/ynnickw/applaunchflow-mcp) for template-based layout generation (device mockups, headlines, backgrounds) - **8 Flutter-specific gotchas** documented from real projects (Firestore overwriting seed data, Provider type mismatches, animation settle times, etc.) Everything from the original SwiftUI path is preserved — the 11 SwiftUI gotchas, step-based coordinator, element rendering, and locale looping all work as before. ## What it does - Builds an in-app capture system for **SwiftUI** and **Flutter** iOS apps — zero production footprint - Seeds deterministic demo data so every screenshot looks populated and polished - Navigates to each screen programmatically and snapshots including status bar and safe area - Renders isolated elements (cards, widgets, charts) with transparency - Loops every locale automatically — one build, all languages - Evaluates results and preselects the strongest shots for App Store use - Hands off to AppLaunchFlow MCP for layout generation (optional) ## Install ### Using npx skills (recommended) ```bash npx skills add ynnickw/applaunchflow-skill ``` This works with Claude Code, Cursor, Windsurf, OpenCode, Codex, and [40+ other agents](https://github.com/vercel-labs/skills#available-agents). Install globally (available across all projects): ```bash npx skills add ynnickw/applaunchflow-skill -g ``` Install for a specific agent: ```bash npx skills add ynnickw/applaunchflow-skill -a claude-code ``` ### Manual (git clone) ```bash git clone https://github.com/ynnickw/applaunchflow-skill ~/.claude/skills/applaunchflow-skill ``` ## Usage Once installed, the skill triggers automatically when you ask your agent to: - Capture marketing screenshots for my iOS app - Generate locale screenshots across all languages - Render my widgets as isolated PNGs - Automate App Store screenshot capture Or just tell the agent what you need: ``` > Capture marketing screenshots for my app across all locales ``` The agent explores your codebase, detects screens/locales/devices automatically, and proceeds. It only asks when something is genuinely ambiguous. ## Example prompts ### Breathwork app (Flutter) ```text Capture marketing screenshots for my Flutter breathing app. I want all 4 tabs: Home, Routine, Library, and Progress. Also capture the technique detail screen and the breathing session player mid-session. English and German, iPhone 17 simulator. ``` ### Coffee app (SwiftUI) ```text Capture marketing screenshots for my coffee tracking app. I want Home, Shelf, Coffee Detail, Brew Timer, and Settings. Also render coffee cards and all my widgets as isolated elements. Capture across en, de, es, fr, ja. ``` ### Minimal prompt ```text Capture marketing screenshots for my app ``` The skill will auto-detect everything and proceed. ## Output ``` marketing/ en/ 01-home.png 02-detail.png 03-settings.png elements/ card-morning-blend.png widget-pulse-small.png de/ ... ``` After capture, the skill evaluates all full-screen shots and produces a shortlist of up to 10 App-Store-ready candidates. If the AppLaunchFlow MCP is configured, it uploads and generates layouts automatically. ## Requirements ### SwiftUI - Xcode 16+ - iOS 17+ deployment target - A simulator runtime matching your target iOS version - Python 3 ### Flutter - Flutter SDK 3.10+ - Xcode (for iOS simulator) - A booted iOS simulator - Python 3 ## Credits Forked from [ParthJadhav/ios-marketing-capture](https://github.com/ParthJadhav/ios-marketing-capture), originally built for [Bloom](https://apps.apple.com/us/app/bloom-coffee-shelf-recipe/id6759914524). ## License MIT

applaunchflow

Installation

SKILL.md