explore
코드베이스 탐색 - 서브에이전트에서 실행하여 토큰 절약
Discover development debugging skills, compare source repositories and read the instructions for each workflow.
코드베이스 탐색 - 서브에이전트에서 실행하여 토큰 절약
Dual-layer caching strategies for the Flare Stack Blog. Use when implementing CDN cache headers, KV caching with versioned invalidation, or debugging cache-related issues.
Testing architecture and best practices for the Flare Stack Blog. Use when writing, debugging, or refactoring tests with Vitest and Cloudflare Workers pool.
Reviews code changes for bugs, style issues, and best practices. Use when reviewing PRs, checking code quality, or validating changes before commit.
Analyze system processes and resource usage to diagnose runaway processes. Use when investigating CPU/memory/IO spikes.
Analyze performance of macOS/iOS apps with Instruments/xctrace (Time Profiler, allocations). Use when the user needs profiling or stack analysis for native apps.
Web/React interrogation using HAR + Playwright trace and endpoint mapping (behavior-first when sourcemaps absent).
Analyze and report authorized evidence using Recon Workbench (rwb) workflows. Use when you need authorize/plan/run/summarize flows and evidence-backed reporting for web apps or OSS repos.
Static triage for macOS .app bundles and Mach-O binaries; bundle map, signature/entitlements, imports/symbol hints.
iOS Simulator interrogation: simctl screenshots/video, Web Inspector for web content, Instruments for native behaviors.
Advanced codebase search using Mantic for finding files, understanding context, and analyzing code impact. Use this skill when searching for code, files, or understanding code relationships.
Enterprise-grade repository analysis with arc42/C4 architecture documentation, technical debt quantification, security assessment, and multi-stakeholder reporting
Profiling, optimization techniques, and performance best practices
Check if a Ralph run is active and tail the most recent log.
Identify friction points, bottlenecks, bugs, and technical debt. Use for audits, debugging sessions, when something feels wrong, or before major work to surface hidden issues. This is the second system in the 5...
Explains code in beginner-friendly terms. Use when user asks 'what does this do', 'how does this work', 'explain this code', or needs to understand generated code. Breaks down complex code into simple explanati...
Compares Figma screenshots with HTML screenshots using pixel-level analysis. Use when verifying visual accuracy of Figma-to-HTML conversion.
Documents screen-to-screen navigation flows, user journeys, and state transitions across multiple screens. Updates the "画面フロー" section in screen spec.md.
Systematic debugging discipline that enforces root cause investigation BEFORE fixes. Triggers on - bug, error, failing, broken, fix, not working, crash, issue, debug, investigate, diagnose, trace, root cause, w...
Use pre-computed discovery files at .claude/state/ for codebase inventory questions. Triggers on component, hook, schema, form, type, inventory, what exists, list all, how many. Saves 90%+ tokens vs source scan...
Diagnose and resolve build failures, performance issues, and deployment problems using the RAPID framework. Covers TypeScript/build errors, performance bottlenecks, Edge Functions, environment configs. Parallel...
Systematic backward tracing through call chains to find where bugs originate. Triggers on debug, trace, root cause, call stack, error, exception, bug, investigate, find the source, test pollution, flaky test, b...
Web application testing skill using Antigravity browser and Playwright CLI. Use for E2E testing, visual debugging, interaction testing, accessibility checks, and deep audit strategies. Supports both live browse...
Analyze coding sessions to detect corrections and preferences, then propose targeted improvements to Skills used in the session. Use this skill when the user asks to "learn from this session", "update skills",...
The default order uses GitHub repository stars from our indexed data. Stars belong to the source repository and may be shared by multiple skills. They indicate repository interest, not skill installs or an independent quality score.