Reviews and strengthens implementation plans through adversarial rubber-duck dialogue. Scores completeness, feasibility, scope, testability, risk, and assumptions, then iterates one pointed question at a time to drive every dimension to 5/5, re-scoring after each round. Verifies checkable claims with local evidence. Updates the plan with resolved answers. Use when asked to "review my plan", "rubber duck this", "stress test this plan", "is this plan ready", "challenge my plan", "get this plan to 5/5", "make this plan bulletproof", "drive every dimension to 5/5", "what am I missing", "verify this", "is this true", "prove it", "check this claim", "fact-check", or before starting implementation on a non-trivial plan.
Provides architecture guidance for multi-tenant SaaS platforms on Cloudflare or Vercel. Use when defining domain strategy, tenant identification, isolation, subdomain routing, custom domains, white-label setup, tenant separation, plan/limit mapping, building a multi-tenant application, or asking "how do I support multiple tenants" or "build a white-label platform".
Generates folder structures, module contracts, middleware pipelines, and frontend/backend boundaries for TypeScript full-stack applications, and finds domain-informed deepening opportunities in existing codebases. Use when starting a project, setting up project structure, organizing a monorepo, configuring middleware, defining folder layout, designing backend modules, establishing team conventions, improving the architecture of an existing codebase, finding refactor opportunities, deepening the architecture, or asking "how should I structure this app", "design the folder structure", "set up the architecture", "improve codebase architecture", or "find architecture improvements".
Collaborative interrogation that produces an implementation plan. Reads relevant docs and code first, asks one question at a time with a recommended answer, grills the core decisions, flags fuzzy terminology, and walks the decision tree until shared understanding is reached. Outputs a plan file. Use when asked to "create a plan", "help me think through this", "plan this feature", "I want to build X", "grill me", "grill with docs", "understand the docs", "unpack the decisions", "brainstorm a spec", "what should the plan be", "think this through with me", or before starting any non-trivial implementation.
Produces a read-only review report of the current local diff or branch — it lists findings and does NOT edit files. Use when asked to run `/pr-reviewer` before commit, before push, or before handing changes off for PR creation or update; also use for "review my changes", "code review", or "code quality review". Also use for "thermo-nuclear review", "deep code quality audit", "structural review", "harsh maintainability review", or "code judo" — these load the structural quality rubric for an unusually strict maintainability pass. Also use for "deslop this", "clean up AI code", "remove slop", or "review for AI patterns" — these load the AI slop detection catalog. Also use for "security audit", "find vulnerabilities", "deepsec", "security review", "threat model", or "audit for security" — these switch to a whole-codebase security sweep instead of diff-only. For automatic fix-in-place (no manual review step needed), use the private `simplify` skill instead.
Creates bold, minimal, dark-first presentations with structured narrative arcs, punchy slide copy, high-contrast visual design, and conversational speaker notes. Adapts for live talks or async investor pitch decks. Use when creating a presentation, structuring a deck, writing slides, asking "outline a presentation about...", "write slides for...", "design a deck for...", or building a pitch deck for investors.
Defines visual systems and design direction before code is written — colour palettes, typography scales, layout patterns, design tokens, and component styling for product dashboards (SaaS/admin/data-heavy) or marketing/brand landing pages. Includes CRO strategy, conversion benchmarks, persuasion psychology, A/B testing methodology, and social proof patterns. Use when choosing visual direction, selecting colour palettes and fonts, establishing layout patterns, starting a new design system, theming a web application, optimising landing page conversions, auditing social proof, or asking "make this look good", "make this look premium", "design the UI for", "full-bleed hero", "pick a visual style", "improve conversions", "landing page CRO", or "add social proof". Scope: design decisions and specs (tokens, palettes, CRO plans). For writing actual Tailwind markup or building/refining components from those decisions, use the `ui` skill instead.
Optimises SEO and technical foundations for Next.js App Router apps including sitemaps, meta tags, structured data, canonical URLs, redirects, indexing policy, hreflang/internationalisation, Core Web Vitals, security headers, privacy/consent, and error-page resilience. Use when asked to improve SEO, add sitemap.xml, fix meta tags, add structured data, set canonical URLs, set up redirects, fix soft 404s, add hreflang, add security headers (CSP/HSTS), add cookie consent or a privacy policy, fix error pages, improve Core Web Vitals, audit SEO, or build SEO pages at scale. Performs no visual redesigns.