# Agent Skills — Full Reference

> Agent Skills is an open-source, agent-neutral autonomous product-building OS for AI coding tools: 23 skill categories, 28 specialized agents, and 149 reference docs that turn one-line natural-language prompts into deployed, tested, production-ready products on Cloudflare. The same skill content compiles to 37 platform variants (Cursor, GitHub Copilot, Windsurf, Claude Code, OpenAI Codex, Gemini, Continue, Cline, Aider, and more) — no privileged tool. A skill router loads only the smallest useful subset of skills for each prompt, minimizing context-window usage.

This file is the complete, machine-readable reference: every skill, agent, platform variant, and capability. Version 7.2.1. Source: https://agent.skillsl.ink/

## At a glance

- 23 skill categories
- 28 specialized agents (deep-reasoning / balanced / fast model tiers)
- 149 reference docs (loaded on-demand by the router)
- 37 platform variants (one skill set → every AI coding tool)
- Open source, free forever · Cloudflare-native · deploy + production-E2E on every change

## Main capabilities

### Autonomous Engineering
- One-line prompt → deployed, tested product
- Architect → build → test → deploy → verify, zero hand-holding
- 4-tier approval: autonomous · review · approval · blocked
- 28 specialized agents orchestrated in parallel
- Self-improving — folds lessons back into the skills
- Self-healing — Sentry alert → agent traces the error → auto-merging fix PR

### Cloudflare-Native Full Stack
- Workers + Hono edge APIs, Zod at every boundary
- D1 / Neon + Drizzle ORM · R2 · KV · Durable Objects · Queues
- Workflows · AI Search + Vectorize RAG · Workers AI + AI Gateway
- Clerk auth · Stripe + Link payments · Amazon SES email
- React 19 + Vite + shadcn/ui or Angular 22 + Spartan UI

### Quality & Verification
- TDD-first: a failing Playwright E2E before any code
- Real-browser journeys × 6 breakpoints, zero console errors
- AI visual QA · axe WCAG 2.2 AA · Lighthouse ≥ 95
- Deploy + production E2E on every single change
- Feature flags · contract-first AI · drift detection

### Cinematic Websites
- One line → gorgeous, cinematic, functional site
- Competitor research — outscore every peer by ≥ 15%
- Maximalist enrichment + AI-native spiral
- SEO / GEO · JSON-LD · PWA · Core Web Vitals budgets
- Generative media: Ideogram 4.0 · GPT Image 2 · Veo 3.1

### Runs Everywhere
- 37 platform variants from one skill set
- Claude Code · Codex · Cursor · Copilot · OpenCode · Windsurf · Gemini
- Per-platform adapters preserve precision
- Agent-neutral core — no privileged tool
- MCP-native · browser-agentic · edge-first

### Intelligence & Growth
- Remembers & evolves your preferences — confidence-scored Voice-of-Customer model
- Autonomous idea engine — evidence-backed improvement proposals, like an internal co-founder
- Observability from day one — PostHog analytics + Sentry error tracking + feature flags
- Conversion optimization + A/B testing + growth instrumentation
- Cost-aware routing — spends promo credits before metered API

### Trust & Compliance
- Legal pages + GDPR / CCPA / ADA Title II (WCAG 2.1 AA) compliance
- Security hardening — CSP Level 3 + Trusted Types + OWASP Top 10
- Email deliverability — SPF / DKIM / DMARC, bulk-sender compliant
- Every feature dark-launched behind a flag (0% rollout → promote)
- Contract-first AI — Zod-validated typed outputs at every boundary

## Skill categories (23)

Each category holds specialized submodules + reference docs; the router loads only the relevant ones per prompt.

- **01 Operating System** — Supreme policy layer. Autonomy, speed standards, cross-skill coordination, conflict resolution. (6 reference docs)
- **02 Goal & Brief** — Project thesis before code. Product type inference, user identification, business model. (core skill)
- **04 Preference & Memory** — User preferences with confidence levels. Voice of the Customer model, auto memory system. (3 reference docs)
- **05 Architecture & Stack** — Cloudflare-first platform selection. Workers, D1, R2, KV, Durable Objects, Queues, Vectorize, AI Search. (25 reference docs)
- **06 Build & Slice Loop** — Vertical slices, homepage-first. Anti-placeholder rules — real content, real interactions. (26 reference docs)
- **07 Quality & Verification** — 5-level verification pyramid. Playwright E2E, AI visual QA, agentic security testing. (27 reference docs)
- **08 Deploy & Runtime** — Deploy after every change. Typecheck → deploy → purge → E2E → visual verify → fix-forward. (11 reference docs)
- **09 Brand & Content** — Copy system, headline rules, trust surfaces, SEO, structured data, anti-AI-slop filters. (8 reference docs)
- **10 Experience & Design** — Anti-AI-slop design system. Dark-first, bold typography, cascade layers, container queries. (3 reference docs)
- **11 Motion & Interaction** — Meaning-first animation. Scroll-driven, View Transitions API, prefers-reduced-motion mandatory. (1 reference doc)
- **12 Media Orchestration** — Image generation, logo/icon sets, video, social previews, OG images, compression pipelines. (12 reference docs)
- **13 Observability & Growth** — PostHog analytics, GA4 via GTM, Sentry, Stripe billing, feature flags, conversion optimization. (11 reference docs)
- **14 Idea Engine** — Autonomous internal co-founder. Evidence-backed improvements, self-critique filter, auto-implement. (core skill)
- **15 Site Generation** — Research-saturated, maximalist site builds. Competitor-beating, AI-native, deploy-verified. (16 reference docs)
- **16 Cinematic Prime Directive** — One-line prompt to 100 build-breaking rules across 10 quality categories before DONE. (core skill)
- **21 App Foundation** — Non-inferable business requirements plus the exact stack and brand choices every build must satisfy. (core skill)
- **22 Visual Experience** — The cinematic UI bar: anti-slop premium, brand tokens, motion, logo and image quality, WCAG 2.2 visuals. (core skill)
- **23 Content & SEO** — Real-brand copy, per-page SEO and JSON-LD, pSEO, GEO/AI search, citations, i18n, trust surfaces. (core skill)
- **24 Functional** — What every app must DO: complete features, working forms, notifications, flags, embarrassingly easy UX. (core skill)
- **25 Quality & Performance** — Core Web Vitals, WCAG 2.2 AA, TDD-first real-browser E2E, and the build-fail quality gates. (core skill)
- **26 Platform & Delivery** — Cloudflare-first hosting, prod-only discipline, deploy → prod-verify loop, atomic deploys and rollback. (core skill)
- **27 Trust & Compliance** — Legal pages, PII and deletion rights, CSP, Trusted Types, Zod boundaries, AI-agent security, RFC 7807 errors. (core skill)
- **28 CF Integrations Reference** — Consolidated Cloudflare and integrations reference: platform, data, AI edge, media, payments, site generation. (core skill)

## Specialized agents (28)

Each agent declares a model tier (deep reasoning / balanced / fast) that maps to whatever models your AI tool provides. The meta-orchestrator spawns them in parallel.

### Deep reasoning tier (8)
- **Architect** — Analyzes project structure, generates repo maps, designs task graphs, identifies architectural seams.
- **Completeness Checker** — Verifies nothing was missed. Feature Completeness Engine, Zero Recommendations Gate, visual verification.
- **Security Reviewer** — OWASP Top 10, secrets exposure, injection flaws, auth bypasses, CSP issues. Read-only — never modifies code.
- **Visual QA** — Screenshots at all breakpoints. AI vision detects layout breaks, misalignment, text overflow, broken images.
- **Computer Use** — Desktop automation via Computer Use MCP. Finder, System Settings, native apps, cross-app workflows.
- **Meta Orchestrator** — Master agent. Knows every tool, MCP, and skill. Plans multi-system workflows, chains MCPs, spawns agents.
- **Performance Profiler** — Lighthouse audits, Core Web Vitals analysis. Targets LCP≤2.0s, CLS≤0.05, INP≤100ms with specific fixes.
- **Incident Responder** — Sentry-triggered. Reads error events, traces to source, proposes fixes, creates branches, opens PRs.

### Balanced tier (10)
- **Deploy Verifier** — Post-deploy smoke tests. Checks console errors, screenshots 6 breakpoints, runs axe-core, validates SEO.
- **Test Writer** — TDD-first. Writes failing Playwright tests that emulate real users before implementation. Vitest for units.
- **SEO Auditor** — Title, meta, H1, JSON-LD, OG tags, internal links, sitemap, robots.txt. Playwright-powered live audits.
- **Migration Agent** — Drizzle schema migrations. Generates from diffs, validates against D1, tests rollbacks, zero-downtime.
- **Content Writer** — Marketing copy, blog posts, SEO content. Brand voice, Flesch≥60, active voice, zero AI slop.
- **Accessibility Auditor** — axe-core + Playwright. WCAG 2.2 AA audits, reports violations with fix suggestions, verifies remediation.
- **Media Orchestrator** — Creates and optimizes media — images, video clips, TTS audio, podcasts, 3D — coordinating generation pipelines.
- **Motion Choreographer** — View Transitions API, scroll-driven animations, FLIP, staggered reveals. prefers-reduced-motion, always.
- **Browser Operator** — Drives a real browser after every deploy — golden paths, visuals, console, and network verification.
- **Resource Broker** — Normalized registry of owned account resources — credits, quotas, free tiers — with secret-redacted tracking.

### Fast tier (10)
- **Code Simplifier** — Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality.
- **Dependency Auditor** — Scans for outdated deps, security advisories, license violations, unused imports. Prioritized upgrade reports.
- **Cost Estimator** — Estimates Cloudflare Workers costs. Reads wrangler.toml, counts D1 tables, calculates monthly cost.
- **Changelog Generator** — Auto-generates from conventional commits. Parses git log, groups by type, writes user-outcome-focused entries.
- **Changelog Drafter** — Reads git log since the last tag, drafts user-outcome entries grouped by conventional-commit type.
- **Dead Code Remover** — Finds unreachable exports, unused imports, orphaned functions. Confirms zero references, then removes safely.
- **Formatter** — Runs Prettier plus lint autofix on modified files. Consistent formatting with zero human intervention.
- **Model Router** — Recommends the right model tier and effort level for each task before work is dispatched.
- **Renamer** — Semantic rename across the codebase. Greps every reference, updates imports and usages, verifies the result.
- **Transcriber** — Turns a plan step into code, one file at a time. Clean production implementation straight from the spec.

## Platform variants (37)

One skill system, auto-generated convention files for every AI coding tool:

### Modern Formats
Cursor (MDC) · Windsurf · Augment · Copilot Instructions · OpenHands · JetBrains AI · Kiro (AWS) · Void Editor

### Legacy Single-File
.cursorrules · .windsurfrules · .clinerules · .rules · .augment-guidelines · .aider-conventions.md · copilot-instructions.md

### Named Formats
AGENTS.md · GEMINI.md · AMP.md · CODEX.md · QODO.MD · replit.md · CLAUDE.md

### Directory Formats
Amazon Q · Junie · Trae · Tabnine · Kilo · Roo Code · Continue · Codex Skills · Bolt.new · Devin · Goose · OpenCode · Kimi · Claude Plugin · .agents/

## Default stack

Opinionated defaults, escape hatches when you need them:

- **Runtime:** Cloudflare Workers · Hono · Bun
- **Language & Tooling:** TypeScript 7 · oxlint · lefthook
- **Frontend:** React 19 · Angular 22 · Tailwind v4
- **Data:** D1 / Neon · Drizzle ORM · Zod
- **Auth & Billing:** Clerk · Stripe + Link · Turnstile
- **Infra:** Inngest · Amazon SES · Upstash
- **Quality:** Playwright · Vitest · ESLint
- **Observability:** PostHog · Sentry · GA4

## Install

- GitHub Skills: `gh skill install heymegabyte/agent-skills`
- Claude Code plugin: `claude plugin install heymegabyte/agent-skills`
- OpenAI Codex: `git clone https://github.com/heymegabyte/agent-skills ~/.codex/skills/agent-skills`
- Manual (any tool): `git clone https://github.com/heymegabyte/agent-skills`, then use the convention file for your platform

## Links

- [Website](https://agent.skillsl.ink/)
- [GitHub repository](https://github.com/heymegabyte/agent-skills)
- [README (raw)](https://raw.githubusercontent.com/heymegabyte/agent-skills/master/README.md)
- [Changelog (raw)](https://raw.githubusercontent.com/heymegabyte/agent-skills/master/CHANGELOG.md)
- [Short index](https://agent.skillsl.ink/llms.txt)