mahidalhan/claude-hacks

USE WHEN: user requests generative art, algorithmic art, procedural visuals, flow fields, particle systems, creative coding, or says "create art", "generative", "p5.js". DO NOT USE WHEN: static images, traditional graphics, or non-generative visual work is requested. Creates original algorithmic art—never copies existing artists' work.

USE WHEN: analyzing existing architectures, evaluating architectural decisions, identifying technical debt, planning refactoring, assessing whether patterns are justified, onboarding to complex codebases, reviewing PRs with architectural changes. DO NOT USE WHEN: simple code questions, bug fixes, feature implementation, or tasks not requiring architectural analysis. For complex architectural analysis, trigger deep thinking with "think harder" or "ultrathink" to enable thorough first-principles reasoning.

USE WHEN: visualizing algorithms, data structures, architecture, control flow with ASCII art, or when terminal-friendly diagrams are needed. DO NOT USE WHEN: Mermaid diagrams are better suited, rich graphics are available, or simple text explanation suffices.

Explains code, algorithms, system design using ASCII diagrams. Trigger phrases - "explain visually", "I don't get it", "show me", "ascii diagram", "help me understand". Produces diagram-first explanations ending with TL;DR tables.

PROACTIVE skill - Claude invokes automatically when user messages exhibit anti-patterns. Detects recursive nesting, scope creep, buried asks, assumed context, stream dumps, imprecise descriptions. Disable with "coach off" or "raw mode".

USE WHEN: needing real code examples from GitHub, finding API usage patterns, looking up framework setup, verifying library implementations, eliminating hallucinations with working code. DO NOT USE WHEN: working on proprietary code, questions about the current codebase, or when local context is sufficient. Requires Exa API key configured in environment.

USE WHEN: user asks "explain", "teach me", "how does X work", "help me understand", or wants to learn concepts from first principles with interactive comprehension checks. DO NOT USE WHEN: user wants quick answers, code implementation, or already understands the concept and just needs a reminder.

USE WHEN: building web components, pages, or applications requiring distinctive visual design, user says "design", "UI", "frontend", "interface", or wants production-grade aesthetics. DO NOT USE WHEN: backend work, API development, simple scripts, or when generic styling is acceptable.

USE WHEN: user needs diagrams, flowcharts, ERDs, sequence diagrams, state machines, architecture visualizations, or says "visualize", "diagram", "chart", "show me". DO NOT USE WHEN: ASCII art suffices, user wants code not diagrams, or when a simple text description works better.

USE WHEN: user wants to publish, port, or share a skill to a marketplace repo. DO NOT USE WHEN: creating new skills, installing skills, or general questions.

USE WHEN: user asks to create a skill, scaffold a new capability, codify expertise into reusable instructions, or says "create skill", "new skill", "make a skill". DO NOT USE WHEN: user wants to use existing skills, needs general help, or is asking about skill concepts rather than creating one.

Swift Concurrency review and remediation for Swift 6.2+. Use when asked to review Swift Concurrency usage, improve concurrency compliance, or fix Swift concurrency compiler errors in a feature or file.

USE WHEN: planning complex multi-step tasks, organizing work with dependencies, identifying parallelization opportunities, breaking down large features into atomic tasks. DO NOT USE WHEN: single simple tasks, tasks without dependencies, or when user just wants immediate execution without planning. For complex orchestration, trigger deep thinking with "think harder" or "ultrathink" to map comprehensive dependency graphs.

USE WHEN: teaching complex topics requiring chunked delivery, handling mid-explanation questions, preserving context across interrupts, deep-dive explanations with nested topics. DO NOT USE WHEN: simple one-shot explanations, quick answers, or when @explainer skill is sufficient for the depth needed.

USE WHEN: writing system prompts, structuring complex instructions, engineering context for Claude agents, optimizing attention efficiency in long prompts. DO NOT USE WHEN: writing regular code, simple messages, or content where Markdown suffices. For optimal context architecture, trigger deep thinking with "think harder" or "ultrathink" to analyze attention economics thoroughly.