skills

Review or improve code — one skill, four modes: bug/security review (default), performance, clean-code audit, slop cleanup. Pass mode:review|perf|clean|cleanup or infer. Don't use for writing features or generating tests (use test-coverage).

Generate professional diagrams as valid draw.io XML — flowcharts, architecture, C4 models, ER diagrams, sequence diagrams, mind maps, and swimlanes. Don't use for Excalidraw or Mermaid output, hand-drawn sketch styles, or slide decks/presentations.

Generate diagrams as valid Excalidraw JSON — flowcharts, architecture, ER diagrams, mind maps, sequence diagrams, wireframes, C4 models, and more. Don't use for draw.io/Mermaid output, slide decks, or pixel-perfect brand graphics.

Generate a diagram and route to the right engine — draw.io XML (precise, editable, C4, swimlanes) or Excalidraw JSON (hand-drawn, sketch, wireframes). One entry for flowcharts, architecture, ER, sequence, mind maps. Don't use for Mermaid or slides.

Generate landing pages with PAS, AIDA, or StoryBrand — marketing copy from a product brief, or a scannable landing-page rewrite of an existing README. The one skill for landing-page work. Not for blog posts, UX audits, or HTML builds.

Review a product codebase and landing page against 32 viral principles and produce a Virality Score plus ranked fixes. Use to audit virality or prioritize growth. Don't use for SEO, ASO, copywriting, or code review.

Review UI for usability issues using Steve Krug's principles and produce a scannable report. Use when asked for a usability audit, UX review, or UI feedback on screenshots, URLs, or code. Don't use for visual/brand design critique, accessibility (WCAG) audits, or backend/API review.

Sync a GitHub fork with upstream while keeping unmerged feature branches and open PRs mergeable. Use for rebase on upstream, PR conflicts, integration main. Don't use for git init, releases, or non-fork repos.

Manage AI agents in tmux: spawn, status/inspect, message, read replies, or kill sessions via send-keys/capture-pane. Use to launch fleets or talk to running agents. Don't use for SSH, GNU screen, or GUI apps.

Create or update CLAUDE.md and AGENTS.md files following official best practices. Use when asked to create, audit, or improve agent config files (CLAUDE.md, AGENTS.md). Don't use for README/contributor docs or non-Claude IDE plugins.

Optimize App Store and Google Play listings via a 7-phase, plan-approval-gated ASO workflow (keyword strategy, metadata, localization). Don't use for web SEO, paid UA campaigns, or pre-submission compliance audits.

Generate a commit message, stage all changes, and push to remote after scanning for secrets, large files, and protected-branch risks. Skip for opening PRs, code review, or cutting releases/tags.

Check product and brand names for conflicts across trademarks, domains, social handles, and package registries. Returns a risk level and Proceed/Modify/Abandon recommendation. Skip for name brainstorming, logo design, or trademark filings.

Build a production-quality CLI tool for any module or application. Auto-detects language, recommends CLI libraries, and follows a 5-step approval-gated workflow: Analyze, Design, Plan, Execute, Summarize. Don't use for building GUI/TUI apps, web APIs, or authoring one-off shell scripts.

Configure pre-commit hooks and lean GitHub Actions for shift-left quality assurance. Use when adding or auditing CI/CD to maximize local test coverage and minimize CI cost. Skip for Terraform/K8s, deployment pipelines, or non-GitHub CI providers.

Generate and restructure project docs into a clear, accessible hierarchy. Use to organize, generate, or restructure a README. Don't use for API reference generation (JSDoc/Sphinx), landing pages, or agent-config files like CLAUDE.md.

Build production-grade frontend interfaces with distinctive aesthetics and working code. Use when asked to create a UI, component, page, or frontend feature. Skip for backend/API work or usability-only audits.

Validate app/startup ideas with market, feasibility, commercial, and open-source competitor analysis. Use when asked to evaluate, validate, or score a product idea. Don't use for PRDs, go-to-market plans, or investor decks.

Generate cross-platform install scripts for any software or library — a standalone install.sh runnable via a curl/wget one-liner with OS, arch, and package manager detection. Don't use for Dockerfiles, CI/CD pipelines, or one-off shell scripts.

Generate professional SVG logos from project context, producing 7 brand variants (mark, full, wordmark, icon, favicon, white, black) plus a showcase HTML page. Skip for raster-only logos, product illustrations, or full brand-guideline docs.

Optimize Ollama configuration for the current machine's hardware. Use when asked to speed up Ollama, tune local LLM performance, or pick models that fit available GPU/RAM. Don't use for LM Studio, llama.cpp, vLLM, or hosted-API LLM providers.

Run coding tasks via opencode using free cloud models. Use when asked to offload work to opencode.ai or run a free model. Don't use for local models (Ollama, LM Studio), Claude/OpenAI calls, or when Claude should do the work itself.

Transform a project into a professional open-source repository by adding LICENSE, README, CONTRIBUTING, CODE_OF_CONDUCT, SECURITY, and GitHub issue/PR templates. Don't use for documentation overhauls, landing-page generation, or registry publishing.

Generate Product Requirements Documents from `idea.md` and `validate.md` files. Use when asked to create or update a PRD. Don't use for TAD, sprint tasks, or raw idea validation.

Manage software releases end-to-end: bump version, generate changelog, tag, push, GitHub release, publish to PyPI/npm. Use when asked to ship, cut a release, or tag a version. Don't use for routine commits or marketplace publishing.

Install local-first security hardening: pre-commit secret detection, offline dependency scans, static analysis, reports, and gated free CI. Use when hardening repos or adding security hooks. Don't use for incident response or cloud security reviews.

Audit and optimize websites for technical SEO, content SEO, and AI bot accessibility. Fixes meta tags, sitemaps, robots.txt, structured data, llms.txt, and GPTBot/ClaudeBot directives. Not for App Store ASO, paid search, or blog writing.

Create, evaluate, or improve Claude Code subagent files (.claude/agents/*.md) — the frontmatter + system prompt defining a delegatable specialist. Don't use for skills (skill-creator), CLAUDE.md/AGENTS.md (agent-config), or running an agent.

Generate a Technical Architecture Document (TAD) from a PRD. Use when asked to design system architecture or define how a product is built. Updates tad.md and reports GitHub links. Don't use for PRD authoring, sprint tasks, or code implementation.

Generate development tasks from a PRD file with sprint-based planning. Use when users ask to create tasks from PRD, break down the PRD, generate sprint tasks, or want to convert product requirements into actionable development tasks. Creates/updates tasks.md and always reports GitHub links to changed files. Don't use for writing a PRD, authoring a TAD, or executing tasks (see openspec-task-loop).

Generate unit tests for untested branches and edge cases. Use when coverage is low, CI flags gaps, or a release needs hardening. Not for integration/E2E suites, framework migrations, or fixing production bugs.

Build an improved website clone from a URL via 6-phase gated workflow (Vite/React/shadcn/Tailwind + GitHub Pages). Use for end-to-end site rebuilds. Don't use for single-phase work.

Analyze any website URL across 6 dimensions: UI/UX, category, style, performance (LCP/CLS/TTFB/weight/requests), surface-level security, and SEO (score 0-100 with breakdown). Outputs structured JSON for downstream skills. Use when asked to analyze, audit, or scan a website. Don't use for full SEO audits or penetration testing.

Execute the approved tasks.md to build a Vite + React + shadcn/ui + Tailwind CSS website deployable to GitHub Pages. Collects assets from the original site and creates new assets per plan. Emits builder metadata for the final report. Use when asked to build, implement, or code a website from a plan. Don't use for design review or planning — those are upstream phases.

Convert website analyzer output into a comprehensive plain-language report for non-technical users. Prompts for validation before persisting. Use when asked to report on a website analysis, create an end-user report, or translate technical metrics into plain language. Don't use for technical audit reports targeting developers, raw SEO audits, or penetration testing. Approval gate: never saves without explicit user approval.

Turn approved end-user report into a full improvement proposal with what/why/value for each change. Writes prd.md after user approval. Use when asked to propose website improvements, create a PRD for a site rebuild, or plan improvements with measurable impact. Don't use for implementation planning or coding — those are separate phases.

Produce a before/after comparison report of a website clone project. Uses Phase 1 analysis as baseline and builder metadata as 'after' snapshot. Covers performance, SEO, security, UI/UX deltas, deviations, and GitHub Pages URL. Use when asked for a final report, before-after comparison, or project closure summary of a website clone. Don't use for ongoing monitoring or live site audits.

Turn approved prd.md into a phased implementation plan with landing page first, asset collection vs creation, individual tasks. Writes tasks.md after user approval. Use when asked to plan website implementation, create an implementation plan, or break down a PRD into tasks. Don't use for building or coding — that's a separate phase.

Audit iOS/macOS app projects against Apple App Store Review Guidelines to catch rejection risks before submission, with per-guideline PASS/FAIL/WARNING verdicts and fix suggestions. Don't use for Google Play/Android submissions, general code review, or post-rejection appeal drafting.