CosmicAtlasPacker

[Utility] Analyzes user prompts and efficiently loads only relevant rule documents. Auto-generates category-document mapping config file on first invocation by scanning project documents. Supports both automatic and manual invocation. (project)

[Dev] Extracts and organizes best practices for a given topic into a minimal tree structure (max depth 3, max 5 children per node). Use during task planning when writing subtasks in Docs/{name}_Task.md - output is added under each subtask as a concise reference guide. Pure reasoning task with strict formatting rules: keywords/noun phrases only, no prose. (project)

[Utility] Analyzes user prompts and efficiently loads only relevant rule documents. Auto-generates category-document mapping config file on first invocation by scanning project documents. Supports both automatic and manual invocation. (project)

[Design System] Quantitative accessibility audit for UI - contrast ratios, font sizes, tap targets, heading hierarchy. Use when (1) checking WCAG color contrast compliance, (2) auditing text sizes for readability, (3) validating touch/click target sizes, (4) reviewing heading structure and landmarks, (5) user asks to 'check accessibility', 'audit contrast', 'WCAG compliance', or 'a11y check'.

[Aesthetics] Positions aesthetics within art/design history, identifies what traditions it echoes, and spots clichés, generic trends, and AI-slop patterns to avoid. Outputs Critical Notes as Memo nodes in Brain canvas.

[Aesthetics] Maps aesthetic briefs to cultural references: art movements, photography, film, architecture, music, fashion. Outputs reference lists with visual language analysis. Organizes findings as Image and Memo nodes in Brain canvas.

[Aesthetics] Analyzes form, composition, and spatial relationships: eye paths, balance/asymmetry, rhythm, layering, fore/mid/background usage. Outputs compositional rules as Memo nodes in Brain canvas.

[Aesthetics] Defines temporal aesthetics: speed, acceleration habits, tempo, static vs dynamic balance, motion weight and inertia. Produces Motion Bible as Memo nodes in Brain canvas.

[Aesthetics] Extracts recurring visual patterns from references: contrast habits, shapes, density/whitespace balance, rhythm, textures. Produces Aesthetic Pattern Library as Data-Sheet nodes in Brain canvas.

[Utility] ⛔ MANDATORY skill for ANY Brain canvas operations. AI MUST invoke this skill BEFORE creating/modifying ANY files in Docs/Brain/. Failure to follow = broken canvas.

[Design System] Lightweight brand guidelines enforcement for UI copy and visual motifs. Use when (1) checking UI labels/buttons/error messages against brand tone, (2) validating color usage in specific contexts, (3) ensuring reserved components are used correctly, (4) user asks to 'check brand guidelines', 'validate brand compliance', 'review copy tone', or 'enforce design rules'.

[Design System] Validate that code and styles use official design tokens instead of hard-coded values. Use when (1) reviewing CSS/SCSS/Tailwind/styled-components for hard-coded colors/spacing, (2) checking SwiftUI/UIKit for raw color/font values, (3) auditing Unity styles for magic numbers, (4) enforcing design system compliance, (5) user asks to 'check design tokens', 'validate tokens', 'find hard-coded values', or 'audit design system usage'.

[Flame] Flame engine quick reference. Component lifecycle, Collision, Effects, Camera and core API reference. (project)

[Flutter] Flutter code obfuscation guide. Build commands, symbol files, stack trace deobfuscation, and limitations. (project)

[Flutter] Flutter performance best practices and optimization guide. Build cost, rendering, lists, animations, and anti-patterns. (project)

[Flutter] pub.dev package search skill. Quick package search, info lookup, version check, and dependency analysis. (project)

[UI/UX] Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.

[Task Mgmt] A Skill that logs completed tasks to Docs/History/History_yyMMdd.md as a Markdown table in Korean. Use this when a task or subtask has been completed and the user wants to record the completion time, task title, and a short Korean retrospective line into the History file for that date. (user)

[Utility] Guide for developing Claude Code hooks - shell commands that execute automatically in response to tool calls and notifications. Use this when users want to create, debug, or optimize hooks for automating workflows like linting, formatting, notifications, or custom validations. Supports PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification events.

[Design System] Validate margin/padding/gap values against spacing scale and grid rules. Use when (1) checking if spacing values follow the design system scale, (2) finding off-scale or inconsistent spacing, (3) auditing layout consistency across components, (4) user asks to 'check spacing', 'validate layout', 'audit margins/padding', or 'find off-scale values'.

[Code Quality] Identifies architectural hotspots: high-churn files, coupling issues, layer violations, and dependency tangles. Use to find systemic issues that cause repeated bugs or slow development.

[Code Quality] Generates summary report of refactoring changes for review. Use after completing a refactoring session to document what changed, why, and what to verify.

[Code Quality] Removes code duplication and cleans up dead code. Use to eliminate copy-paste code, consolidate similar logic, and remove unused code paths.

[Code Quality] Extracts methods, classes, or modules to reduce complexity and improve isolation. Use when functions are too long, classes have too many responsibilities, or concerns are mixed.

[Code Quality] Plans Git commit strategy for refactoring: branch naming, commit granularity, commit messages, and safe merge approach. Use to structure version control for reversible, reviewable changes.

[Code Quality] Performs systematic renaming to improve code clarity: variables, functions, classes, files. Use when names are unclear, misleading, or inconsistent with their purpose.

[Code Quality] Evaluates risk level of proposed refactoring changes. Use to assess dependencies, blast radius, rollback difficulty, and determine if changes are safe to proceed.

[Code Quality] Scans codebase for technical debt indicators: code smells, complexity hotspots, duplications, outdated patterns, and anti-patterns. Use when starting a refactoring session to identify what needs attention.

[Code Quality] Defines clear boundaries and goals for a refactoring session. Use to establish what WILL and WILL NOT change, set success criteria, and create a focused refactoring scope.

[Code Quality] Simplifies complex conditionals: deep nesting, long if-else chains, switch sprawl. Use when control flow is hard to follow or has high cyclomatic complexity.

[Code Quality] Establishes test coverage requirements before refactoring. Use to identify missing tests, create minimal safety net tests, and define verification strategy for behavior preservation.

[Utility] Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

[Debug] Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes - four-phase framework (root cause investigation, pattern analysis, hypothesis testing, implementation) that ensures understanding before attempting solutions. (project)

[Task Mgmt] A Skill that adds new tasks with priority tags to Docs/Task/{StepName}_Task.md documents for multi-agent parallel development. AI agent MUST invoke this skill AUTOMATICALLY when (1) discovering new work items during implementation, (2) user requests a new feature/fix, (3) breaking down a large task into subtasks, (4) identifying follow-up work after completing a task. CRITICAL - Never modify document structure, only add tasks with proper priority tags (!high, !medium, !low). Tasks should be distributed to avoid file conflicts between workers. (user)

[Task Mgmt] A Skill that syncs task status in Docs/Task/{StepName}_Task.md by moving tasks between sections (Backlog, Worker1, Worker2, Worker3, Review, Done) with priority-aware management for multi-agent parallel development. AI agent MUST invoke this skill AUTOMATICALLY when (1) starting work on a task - move to assigned Worker, (2) AI completing work - move to Review (NOT Done). CRITICAL - AI CANNOT move to Done or Backlog. Only User/Reviewer can move tasks to Done or Backlog. (user)

[Task Mgmt] A Skill that segments subtasks into more granular, actionable items before starting work. AI agent MUST invoke this skill AUTOMATICALLY when moving a task to In Progress. Break down existing 2-space indented subtasks into smaller, more specific subtasks to provide clearer objectives for the agent. (user)

[Task Mgmt] Verification tool for validating task completion in Review section. Use when checking if tasks are actually done before moving from Review to Done. Verifies subtask completion, code/file changes, and task-work alignment. Optional but recommended for quality assurance. (user)

[UI/UX] Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.

[UI/UX] Visualizes UI screen structures and components as ASCII art. Represents wireframes, layouts, and UI components as text-based diagrams. Use when requesting 'ASCII wireframe', 'screen structure visualization', or 'UI layout drawing'.

[UI/UX] Visualizes user flows and screen transitions as ASCII diagrams. Represents navigation flows, user journeys, and screen-to-screen paths. Use when requesting 'flow diagram', 'user journey visualization', or 'navigation flow'.