{
  "markdown": "<div align=\"center\">\n\n# 🚀 Ultra Skill\n\n**The Ultimate AI Coding Super-Skill**\n\n[![GitHub stars](https://img.shields.io/github/stars/amirudinR/Ultra-Skill?style=for-the-badge&color=7c3aed&logo=github)](https://github.com/amirudinR/Ultra-Skill)\n[![License: MIT](https://img.shields.io/badge/License-MIT-7c3aed?style=for-the-badge)](LICENSE)\n[![Version](https://img.shields.io/github/v/release/amirudinR/Ultra-Skill?style=for-the-badge&color=7c3aed)](https://github.com/amirudinR/Ultra-Skill/releases)\n[![Platforms](https://img.shields.io/badge/Platforms-11%2B-7c3aed?style=for-the-badge)]()\n[![Skills](https://img.shields.io/badge/Skills-40%2B-7c3aed?style=for-the-badge)]()\n[![Slash Commands](https://img.shields.io/badge/Commands-14-7c3aed?style=for-the-badge)]()\n\n**The first AI coding skill that covers the entire SDLC — from idea to shipped product, with YAGNI philosophy, TDD discipline, and design intelligence built in.**\n\n</div>\n\n---\n\n## 📖 What Is Ultra Skill?\n\nUltra Skill is the **first unified AI coding super-skill** — a complete, professionally-crafted system that guides AI coding agents through the **full software engineering lifecycle**, from idea to shipped product. It combines battle-tested workflows, strict quality gates, a lazy senior dev philosophy, and an AI-powered design intelligence engine into one cohesive, plug-and-play skill collection.\n\nUnlike fragmented toolkits that cover only one phase of development, Ultra Skill delivers **everything** you need:\n\n| What You Get | Details |\n|---|---|\n| 🧠 **40+ Skills** | Spec, plan, build, test, debug, review, ship, design — every phase covered |\n| 🔧 **14 Slash Commands** | One-command access to every workflow phase |\n| 🤖 **11+ Platforms** | Claude Code, Cursor, Windsurf, OpenCode, Gemini, Copilot, and more |\n| ⚙️ **3 Iron Laws** | Non-negotiable quality gates enforced at every step |\n| 🐴 **YAGNI Philosophy** | Never write code you don't need |\n| 🎨 **Design Engine** | 67 UI styles, 161 color palettes, 22 tech stacks |\n| 👥 **4 Agent Personas** | Parallel review with dedicated specialists |\n\n---\n\n## ✨ Features\n\n| Category | What You Get |\n|---|---|\n| 🧠 **40+ Skills** | 11 categories covering the entire SDLC — spec, plan, build, test, debug, review, ship, and beyond |\n| 🔧 **14 Slash Commands** | One-command access to every workflow phase: `/spec`, `/plan`, `/build`, `/test`, `/review`, `/ship`, and more |\n| 🤖 **11+ Platform Support** | Claude Code, Cursor, Windsurf, Gemini CLI, OpenCode, Cursor, GitHub Copilot, Devin, Cline, Codex, Kiro, Antigravity, and more |\n| ⚙️ **3 Iron Laws** | TDD, Verification, and Debugging — non-negotiable quality gates enforced at every step |\n| 🐴 **7-Rung Ponytail Ladder** | YAGNI-first approach: ask \"does this need to exist?\" before writing a single line |\n| 🏗️ **SPEC→PLAN→BUILD→VERIFY→REVIEW→SHIP** | A structured lifecycle that turns chaos into a repeatable delivery pipeline |\n| 🎨 **Design Intelligence Engine** | 67 UI styles, 161 color palettes, 57 font pairings, 22 tech stacks — AI-powered design at your fingertips |\n| 👥 **4 Agent Personas** | `code-reviewer`, `security-auditor`, `test-engineer`, `web-performance-auditor` — parallel review with `/ship` fan-out |\n| 🧩 **Subagent-Driven Dev** | Parallel agent dispatch with two-stage review (spec compliance → code quality) |\n| 📋 **7 Reference Checklists** | Security, performance, accessibility, testing patterns, observability, definition of done, orchestration |\n| 🪝 **Runtime Hooks** | Session activation, mode tracking, status line, and lifecycle hooks for every major platform |\n\n---\n\n## 🗂️ Skill Categories\n\n<details open>\n<summary><b>11 Categories · 40 Skills</b></summary>\n\n| Category | Skills |\n|---|---|\n| 🧠 **meta/** | `ultra-using-skills`, `ultra-writing-skills` |\n| 🐴 **philosophy/** | `ultra-ponytail`, `ultra-yagni-audit`, `ultra-yagni-debt`, `ultra-impact-gain` |\n| 💡 **ideation/** | `ultra-brainstorming`, `ultra-interview-me`, `ultra-spec-driven-dev` |\n| 📋 **planning/** | `ultra-planning`, `ultra-writing-plans` |\n| 🏗️ **implementation/** | `ultra-incremental-impl`, `ultra-subagent-dev`, `ultra-parallel-agents`, `ultra-api-design`, `ultra-context-engineering`, `ultra-source-driven-dev` |\n| 🧪 **testing/** | `ultra-tdd`, `ultra-browser-testing`, `ultra-doubt-driven-dev` |\n| 🐛 **debugging/** | `ultra-debugging`, `ultra-observability` |\n| ✅ **quality/** | `ultra-code-review`, `ultra-code-simplify`, `ultra-performance`, `ultra-security`, `ultra-verify-before-done` |\n| 🔀 **git-workflow/** | `ultra-git-workflow`, `ultra-git-worktrees`, `ultra-finishing-branch` |\n| 📦 **delivery/** | `ultra-ci-cd`, `ultra-shipping`, `ultra-documentation`, `ultra-deprecation` |\n| 🎨 **ui-design/** | `ultra-frontend-eng`, `ultra-ui-styling`, `ultra-design-system`, `ultra-brand-identity`, `ultra-banner-design`, `ultra-slides` |\n\n</details>\n\n---\n\n## 🔧 Installation\n\n### Claude Code (Recommended)\n\n```bash\n# Clone the repository\ngit clone https://github.com/amirudinR/Ultra-Skill.git\n\n# Or add as submodule to your project\ngit submodule add https://github.com/amirudinR/Ultra-Skill.git .ultra\n```\n\n### All Platforms\n\n| Platform | Setup |\n|---|---|\n| **Claude Code** | `.claude-plugin/plugin.json` — Full plugin with session activation, mode tracking, commands, and statusline |\n| **Cursor** | `.cursor/rules/` — Always-on project rule |\n| **Windsurf** | `.windsurf/rules/` — Project rule |\n| **OpenCode** | `.opencode/` — Plugin with skill loading and commands |\n| **Gemini CLI** | `gemini-extension.json` + `.gemini/commands/` — Extension manifest with auto-discovered skills |\n| **GitHub Copilot** | `.github/copilot-instructions.md` — Repository-level instructions |\n| **Kiro** | `.kiro/steering/` — Steering rules |\n| **Antigravity** | `.agents/` — Workspace rules |\n| **Codex CLI** | `.codex-plugin/plugin.json` — Plugin with lifecycle hooks |\n| **Cline** | `.clinerules/` — Project rule |\n| **Devin** | `.devin-plugin/plugin.json` — Plugin manifest |\n\n> For detailed platform-specific guides, see [`docs/platform-setup/`](docs/platform-setup/).\n\n---\n\n## ⚡ Quick Start\n\n```bash\n# Start a new feature with structured spec\nuse skill tool to load ideation/ultra-spec-driven-dev\n\n# Plan before you build\nuse skill tool to load planning/ultra-planning\n\n# Develop with TDD discipline\nuse skill tool to load testing/ultra-tdd\n\n# Debug systematically (root cause, not symptom)\nuse skill tool to load debugging/ultra-debugging\n\n# Review before merge\nuse skill tool to load quality/ultra-code-review\n\n# Ship with confidence\nuse skill tool to load delivery/ultra-shipping\n```\n\n### Slash Commands\n\n| Command | What It Does |\n|---|---|\n| `/spec` | Define requirements with structured spec |\n| `/plan` | Break down tasks and estimate effort |\n| `/build` | Implement with incremental TDD |\n| `/test` | Write and run tests |\n| `/review` | Five-axis code review |\n| `/ship` | Fan-out review + go/no-go decision |\n| `/code-simplify` | Over-engineering cleanup |\n| `/webperf` | Core Web Vitals audit |\n| `/debug` | Root cause investigation |\n| `/ponytail` | Toggle lazy senior dev mode |\n| `/yagni-audit` | Scan for unnecessary code |\n| `/yagni-debt` | Track `ponytail:` shortcuts |\n| `/impact-gain` | Measure refactoring impact |\n| `/brainstorm` | Socratic idea exploration |\n\n---\n\n## 🏛️ Architecture\n\n```\nultra/\n├── ULTRA_SKILL.md              # The one true philosophy — read first\n├── AGENTS.md                    # Agent instructions (portable)\n├── CLAUDE.md                    # Claude Code guide\n├── GEMINI.md                    # Gemini CLI guide\n├── gemini-extension.json        # Gemini extension manifest\n├── opencode.json                # OpenCode plugin config\n│\n├── skills/                      # 📦 40 skills across 11 categories\n│   ├── meta/                     #   Meta: using skills, writing skills\n│   ├── philosophy/               #   Philosophy: ponytail, YAGNI, impact\n│   ├── ideation/                 #   Ideation: brainstorming, spec\n│   ├── planning/                 #   Planning: task breakdown\n│   ├── implementation/           #   Implementation: incremental, subagents\n│   ├── testing/                  #   Testing: TDD, browser, doubt-driven\n│   ├── debugging/                #   Debugging: root cause, observability\n│   ├── quality/                  #   Quality: review, simplify, security\n│   ├── git-workflow/             #   Git: workflow, worktrees, finishing\n│   ├── delivery/                 #   Delivery: CI/CD, shipping, docs\n│   └── ui-design/               #   UI/UX: styles, systems, brand\n│\n├── commands/                    # ⚡ 14 slash commands (.toml + .md)\n├── hooks/                       # 🪝 Runtime hooks for all platforms\n├── agents/                      # 👤 4 agent personas\n├── references/                  # 📋 7 checklists (security, perf, a11y…)\n├── scripts/                     # 🔧 Validation & utility scripts\n├── docs/                        # 📖 Documentation\n└── ultra-design-engine/         # 🎨 Design intelligence (67 styles, 161 palettes…)\n```\n\n---\n\n## 🧘 Philosophy\n\n### The 3 Iron Laws\n\n> **TDD Iron Law** — No production code without a **failing test** first. If code exists without a test, delete it. Start over.\n>\n> **Verification Iron Law** — No completion claims without **fresh verification evidence**. Run tests, read output, then claim.\n>\n> **Debugging Iron Law** — No fixes without **root cause investigation** first. Symptoms are not causes.\n\n### The Ponytail Ladder (7 Rungs)\n\nBefore writing any code, climb this ladder from the top:\n\n```\n1️⃣  Does this need to be built at all?          ← YAGNI\n2️⃣  Does it already exist in this codebase?      ← Reuse\n3️⃣  Does the standard library already do this?    ← Built-in\n4️⃣  Does a native platform feature cover it?      ← Platform\n5️⃣  Does an already-installed dependency solve it? ← Existing\n6️⃣  Can this be one line?                         ← Minimal\n7️⃣  Only then: write minimum code that works      ← Ship\n```\n\n### Code Rules\n\n- ❌ No abstractions not explicitly requested\n- ❌ No new dependency if avoidable\n- ✅ Deletion over addition. Boring over clever\n- ✅ Shortest working diff wins\n- ✅ Bug fix = root cause, not symptom — grep all callers\n- ✅ Non-trivial logic leaves **one runnable check** behind\n\n---\n\n---\n\n## 🤝 Contributing\n\nContributions are welcome! Here's how:\n\n1. **Read `ULTRA_SKILL.md`** — understand the philosophy before contributing\n2. **Read `docs/skills-contributing.md`** — understand the skill anatomy\n3. **Follow the Ponytail Ladder** — ask if your skill is truly needed\n4. **Each skill = one `skills/<category>/<name>/SKILL.md`** — with frontmatter, process, rationalizations, red flags, and verification\n5. **Validate** — run `node scripts/validate-skills.js && node scripts/validate-commands.js` before submitting\n\n### Adding a Skill\n\n```bash\ncp -r skills/template skills/<category>/<your-skill-name>\n# Edit SKILL.md following the skill anatomy\nnode scripts/validate-skills.js\n```\n\n### Adding a Platform\n\nCreate a `.platform/` directory with the appropriate adapter files. See [`docs/agent-portability.md`](docs/agent-portability.md) for the pattern.\n\n---\n\n## 📄 License\n\n[MIT](LICENSE) © Ultra Skill Contributors\n\n---\n\n<div align=\"center\">\n\n**Star ⭐ us on GitHub — it helps other developers discover this project!**\n\n</div>\n",
  "bytes": 11288,
  "sha": "a0bf3873548f35fa5e392a825c17574074ff3c6c2243ebb9f14edcb3cec0c194",
  "repo_slug": "amirudinr/ultra-skill",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_amirudinr_ultra_skill_4e434986/readme"
}