Design e Frontend

Visual design systems, UI/UX styling, Tailwind CSS, CSS variables, component libraries (shadcn/ui, Radix UI), design tokens, accessibility (WCAG), responsive layout, dark mode, and Figma-to-code workflows. NOT for React component logic, hooks, or state management (use react-development). NOT for Next.js routing, SSR, or server components (use nextjs-development).

PM funnel workflow for turning user research into prioritized engineering stories — product story, user research to backlog, user needs to stories, PM workflow. Chains elicitation (deep interviews that surface real needs), user-journey-design (touchpoints, pain points, emotions), persona-definition (who we're building for), outcome-orientation (measurable success criteria), and prioritization (RIC

Um wrapper CAD com foco em diagnósticos para Build123d, oferecendo asserções de interferência/folga em nível de montagem e verificações opcionais de imprimibilidade por peça. Carregue-o antes de editar um `assembly.py` que usa o wrapper ou interpretar seu JSON de diagnóstico; ele é acionado ao executar `khana check`/`build`/`view`/`draw` ou editar um arquivo que importa `cad_kha`.

EXTENDING effective context capacity — KV-cache optimization, observation masking, context partitioning, and retrieval strategies. Use when the user asks to "optimize context", "implement KV-cache", "partition context", "mask observations", or mentions extending context capacity or cache-friendly prompt design. NOT for reducing or compressing content via summarization (use context-compression), NO

Design, evaluate, and iteratively improve prompts for LLMs — system prompts, few-shot examples, reasoning structures, and instruction templates. Use when the user asks to improve a prompt, write a system prompt, optimize LLM instructions, reduce hallucinations through prompt structure, test prompt variants, or apply prompting techniques (structured reasoning, ReAct, few-shot, structured output). N

Apply outcome-over-output product thinking — separating what gets built from what actually changes for the user, running the "so what" test to climb from output to outcome to impact, choosing leading vs lagging metrics, and identifying a North Star metric. Use when the user asks whether something is an outcome or an output in a product context, wants to run the "so what" test on a deliverable, wan

Turns a vague "I want to build a Claude plugin" into a concrete, problem-first plugin idea — or tells you to not build it. Use when you want to build a Claude plugin, have an idea for a plugin and don't know if it's worth pursuing, are asking "should I build this as a plugin or is there already one", are exploring plugin ideas for your team, or are mining your workflow for repeatable pain points t

Stress-test, critique, and challenge existing ideas through pattern recognition, bias detection, Bayesian reasoning, blind-spot exposure, and red-flag identification. Use when the user asks to stress-test a plan, challenge assumptions, find flaws, identify risks in an idea, expose hidden biases, audit a proposal for blind spots, or "poke holes in this". NOT for generating new ideas or solutions (u

Foundational theory of context engineering — what context IS, how attention works, progressive disclosure principles, and context budgeting basics. Use when the user asks to "understand context", "explain context windows", "learn context engineering", or discusses context components, attention mechanics, or context budgets. NOT for fixing broken context or diagnosing failures (use context-degradat

Designs a complete Retrieval-Augmented Generation (RAG) pipeline for a given use case, including chunking strategy, embedding model selection, and retrieval approach.

Design ad-hoc personas for niche domains when the canonical 12 brainstorm- swarm personas don't fit. Covers: when a custom persona is justified (vs forcing canonical to fit), the persona-design template (voice, contribution shape, output format), anti-patterns (too-narrow personas, redundant personas, sock-puppet personas), and how to invoke a custom persona inline via Task() with a tailored promp

Reviews web forms for usability issues — field ordering, validation messages, error states, and accessibility.