skills

Use this skill when the user is starting a new SaaS web app and wants a battle-tested, opinionated stack with the foundational decisions already made — auth, database, billing, transactional email, deploys, AI integration. Captures Next.js App Router + Supabase + Tailwind v4 + Stripe + Resend + Vercel as the core, with optional Gemini / Claude / AI SDK on top. Includes a concrete quickstart checklist, sample RLS migration, Stripe webhook skeleton, and a CLAUDE.md template so a Claude Code agent can work alongside the developer at full velocity from day one. Triggers on "new SaaS starter", "SaaS stack", "Next.js + Supabase starter", "what stack should I use", "ship a SaaS fast", "Tailwind v4 + Supabase", "Stripe subscriptions starter", "Claude Code project setup", "scaffold a SaaS", "modern SaaS boilerplate".

Use this skill when the user needs to build, calibrate, or critique an LLM-based autorater for UX, prose, agent output, or any subjective quality dimension. Covers rubric design, SxS vs SSE selection, anchor writing, the calibration loop, Cohen's kappa / Krippendorff's alpha agreement thresholds, and the anti-patterns that cause judge models to silently disagree with humans. Triggers on "build an autorater", "score this UX", "eval rubric", "LLM-as-judge", "calibrate a judge", "human-AI agreement", "judge model", "rubric design", "side-by-side eval", "SxS vs SSE", "inter-rater reliability", "kappa", "Krippendorff", "autorater calibration", "preference scoring", "scoring rubric".

Use this skill when designing an AI feature that returns UI components (cards, forms, charts, tables, layouts) instead of text. Covers the component-vs-text decision tree, GenUX layout primitives, disclosure rules, validation gates for model-composed UI, and graceful fallback to text when the model can't compose. Triggers on "generative UI", "GenUX", "streamUI", "AI returns components", "AI SDK RSC", "Anthropic Artifacts", "ChatGPT Canvas", "design AI components", "model composes UI".

Use this skill when extracting structured data from images (receipts, menus, statements, IDs, business cards, forms, screenshots) using a vision-capable LLM. Covers schema-first design with zod, validation gates and retry, locale handling (currency / dates / romanization), confidence signalling, and the anti-patterns that cause silent garbage extraction. Triggers on "extract from image", "OCR with structure", "vision model JSON", "Gemini multimodal", "GPT-4 vision extraction", "parse receipt", "parse business card", "image to schema", "structured output from image".

Use this skill when an AI assistant needs to adapt the same factual answer for different audiences (e.g. recruiter vs engineer vs executive, beginner vs expert, customer vs internal). Triggers on "persona-aware", "audience adaptation", "tailor responses", "different personas", "adaptive disclosure", "system prompt branching".

Use this skill when validating a SaaS idea by mining Reddit for real user pain points. Covers the niche → search → score → cluster → brief pipeline, the scoring rubric for separating real pain from noise, sub-selection heuristics, and the anti-patterns that lead to false-positive opportunities. Triggers on "validate a SaaS idea", "mine Reddit", "find pain points", "product discovery", "opportunity research", "user research from Reddit", "competitor pain", "build a brief".