What problem does it solve? Designing competitive mobile apps without evidence leads to guesswork. This Skill gives an agent structured access to the Appllama library of real screens, flows, and UI patterns from top-grossing apps, so research directly informs what gets built. ## Core Features & Use Cases - Tool map and ground rules: Covers all Appllama MCP tools (search_apps, list_app_screens, search_screens, get_screen, boards, credits), plus pagination, expiring media URLs, and rate-limit handling. - Build-from-scratch playbook: A six-phase loop from finding winning apps, studying every screen, and synthesizing category design language to building and verifying each screen in the simulator. - Screen improvement playbook: Diagnose deficits, build a 30+30 reference board via keyword and semantic search, extract a checkable spec, and iterate until the screen matches the best reference. - Use Case: A user asks to build a habit tracker. The agent searches top-grossing habit trackers, walks their onboarding and paywall flows screen by screen, synthesizes the category's patterns, then designs and builds each screen verified against those references. ## Quick Start Use the Appllama MCP to research the top five habit tracker apps and summarize their onboarding and paywall design patterns before we build ours.
