{ "markdown": "

\n \"AI\n \"GitHub\n \"AWS\n

\n\n

\n \"AI\n

\n\n

🧠 Agent Skills Repository

\n\n

\n Created by Yuval Avidani · AI Builder, Speaker, 2× GitHub Star, AWS GenAI Superstar, AI commentator on Channel 12 News (Israel)
\n yuv.ai · all links · @yuvalav · @yuval_770 · YouTube · GitHub\n

\n\n---\n\n## ⚡ Install the full creative stack in one command\n\n```bash\ncurl -sSL https://raw.githubusercontent.com/hoodini/ai-agents-skills/master/install.sh | bash\n```\n\nInstalls **16 skills** as one package: **director** (award-winning film + AI-production Bible, story-first workflow) · **yuv-pilot** (top-of-pyramid orchestrator) · **yuv-design-system** (brand) · **yuv-decks** (cinematic presentations) · **yuv-viral-video** (short-form) · **video-edit** · **video-to-landing-page** · **parallax-landing-page** · **hyperframes** + 4 companions · **nano-banana-pro** (AI images) · **mermaid-diagrams**. Plus ffmpeg / Node / Python / faster-whisper / hyperframes CLI / Python venvs / cross-tool symlinks to ~/.copilot/skills + ~/.agents/skills. Works on macOS, Linux, Windows (Git Bash/WSL).\n\nAfter install, restart your agent and just type what you want:\n\n```\n\"make a deck about for \"\n\"build a stunning parallax landing page from this video: \"\n\"edit this video into a viral short: \"\n```\n\nThe right skill auto-fires. Projects save to **`~/Documents/yuv-projects/{decks,landings,videos}//`** so they're always findable.\n\n> 🌐 **Live examples:** [yuv.ai](https://yuv.ai) · [effects.yuv.ai](https://effects.yuv.ai) (HyperFrames effects catalog) · [examples/parasites/](examples/parasites/) (three parallax landings: GitHub Desktop, Marcus the lion, Hope the cheetah)\n\n---\n\n## 🪜 The YUV.AI skills pyramid\n\nThe YUV.AI creative stack is a **three-tier pyramid**. You don't have to remember which skill does what — say what you want, the orchestrator routes.\n\n```\n ┌─────────────────────┐\n │ yuv-pilot │ ← top: intent → route\n │ (orchestrator) │\n └──────────┬──────────┘\n │\n ┌────────────────────┼────────────────────┐\n ▼ ▼ ▼\n ┌─────────┐ ┌─────────────┐ ┌──────────────┐\n │yuv-decks│ │ yuv-design- │ │ yuv-viral- │\n │ │ │ system │ │ video │\n │ talks · │ │ │ │ │\n │ slides ·│ │ palette · │ │ short MP4s · │\n │ keynotes│ │ typography ·│ │ Reels · TikTok│\n │ │ │ 3 modes · │ │ MrBeast-paced │\n │ │ │ assets │ │ editorial │\n └─────────┘ └─────────────┘ └──────────────┘\n ┌─────────────┐ ┌─────────────────┐ ┌──────────────────┐\n │ video-edit │ │ parallax- │ │ video-to- │\n │ │ │ landing-page │ │ landing-page │\n │ captioned · │ │ │ │ │\n │ Hebrew+EN · │ │ scroll-scrub · │ │ Apple-style · │\n │ transcript- │ │ video frames · │ │ sticky hero · │\n │ review flow │ │ github/lion/hope│ │ frames + sections│\n └─────────────┘ └─────────────────┘ └──────────────────┘\n │\n ▼ (the middle tier uses these)\n open-slide · hyperframes · gsap · nano-banana-pro · video-use ·\n slide-authoring · mermaid-diagrams · hyperframes-cli · website-to-hyperframes\n\n Brand DNA — universal Fly High throughline:\n flight metaphors · HUD + dials · phoenix mark ·\n \"Let's Fly High\" watermark · Anton + JetBrains Mono\n (palette is the chapter, motifs are the brand)\n```\n\n### How to invoke\n\n| Say | What fires |\n|---|---|\n| `\"build me a deck about for \"` | `yuv-decks` (lead) + `yuv-design-system` in **Decks mode** + `open-slide` scaffolding |\n| `\"build a landing page for me / my brand\"` | `yuv-design-system` in **Neon mode** + `gsap` for motion |\n| `\"turn this video into a landing page\"` *(5–15s clip)* | `parallax-landing-page` — scroll-scrub cinematic landing |\n| `\"turn this video into a landing page\"` *(longer, with sections)* | `video-to-landing-page` — Apple-style sticky-hero |\n| `\"edit this video / add captions\"` | `video-edit` — transcript-review-before-render workflow |\n| `\"make it viral / YUV.AI short\"` | `yuv-viral-video` — opinionated MrBeast-paced pipeline |\n| `\"launch my new course / cross-channel campaign\"` | `yuv-pilot` — orchestrates a build order across multiple skills |\n| `\"what should I build for my brand\"` *(strategic, no specific deliverable)* | `yuv-pilot` — surfaces options |\n| `\"build me a snake game\"` *(no YUV.AI signal)* | **Whole pyramid stays out** — picks whatever palette fits |\n| `\"build me a YUV.AI snake game\"` *(explicit brand signal)* | `yuv-design-system` in **Neon mode** + Three.js or Canvas |\n\n### Three palette modes (only when applied)\n\n| Mode | When | Palette |\n|---|---|---|\n| **Neon** *(default for YUV.AI web/app/game/dashboard/social)* | Anything that's not a slide deck | Hot pink `#FF1464` + neon cyan `#00E5FF` + white or rich black |\n| **Decks** *(Fly High)* | Presentations, slides, keynotes only | Purple `#5E17EB` + yellow `#FFEC00` + grey `#F1F2F2` |\n| **Warm Editorial** | Hope, Marcus, bigcats.ai, practical.yuv.ai | Pink + yellow + bone |\n\nThe pyramid auto-applies **only** when YUV.AI brand context is signaled (`my brand`, `for me`, `YUV.AI`, etc.). Generic requests stay outside the system — Claude / Copilot / Cursor pick whatever fits the project.\n\n---\n\n## 🆕 NEW: cinematic-scrub-landing — golden landing page template\n\nDrop in a hero video + logo + product photo. Get back a **premium cinematic landing page** with mouse-driven video scrubbing and a 5-beat narrative arc.\n\n

\n \"Mouse-scrub\n

\n\nThe hero video lives fixed across the whole page — moving the mouse left-right scrubs the timeline so the subject responds to the cursor. Below it, **four fully-opaque sections each carry their own brand identity** (bright primary-accent recipe / cream nostalgia album / dark contemplative pricing / red action finale) and walk the viewer through *longing → joy → nostalgia → contemplation → action*. Works for **any language** (RTL or LTR) and **any subject** — food, tech, animals, fashion, services, SaaS, wildlife, music, books, real estate.\n\n

\n \"Bright\n \"Cream\n \"Dark\n \"Red\n

\n\n**What's baked in (every rule hard-won from a real production build):**\n\n- **All-keyframes ffmpeg re-encode** (`-g 1 -keyint_min 1`) so mouse-scrub is buttery, not stuttery\n- **The z-index trap** documented as hard rule #1 (outer wrapper has NO bg class, or the video disappears)\n- **Per-section opaque identities** — no monotone scrim through every section\n- **Language-aware typography trinity** (Hebrew Rubik+Assistant+Suez One / English Anton+Inter+Caveat / Arabic Cairo+Tajawal+Reem Kufi / Japanese Noto Sans JP / etc.)\n- **RTL/LTR mirroring** — hero panel side, cursor cue side, scroll-to-top position, arrow direction\n- **Six reusable visual primitives** — polaroid, stamp, ribbon, highlight-circle, ghost-text, motif SVGs\n- **Real product photography wins over generated gradients** — the skill always finds prominent placement for `PRODUCT_IMAGE`\n- **Clickable scroll-spy navbar** + scroll-to-top button + favicon set from the logo\n- **12-step verification gate** before reporting done\n\nReference implementation: (Israeli RTL site for the snack \"במבה\" — the original build this skill was distilled from).\n\n🔗 **[View cinematic-scrub-landing →](skills/cinematic-scrub-landing/SKILL.md)**\n\nPairs with `parallax-landing-page` (scroll-scrub frames, page never moves) and `video-to-landing-page` (Apple-style sticky hero with sections below) — pick `cinematic-scrub-landing` when the hero should respond to the **cursor** and the page below should **tell a story**.\n\n---\n\n## 🆕 NEW: yuv-design-system — single source of truth for visuals\n\nYuval Avidani's canonical design system, now public. **Replaces** the prior `yuv-design` skill (which only covered the Warm Editorial pink/yellow/bone family). Encodes:\n\n- **Two palette modes** — **Fly High** (purple `#5E17EB` + yellow `#FFEC00` + grey, the default for keynotes / dashboards / dev-facing UIs) and **Warm Editorial** (pink + yellow + bone, only for the Hope / bigcats / practical.yuv.ai brand family)\n- **Mandatory typography** — Anton + Inter for English, Rubik + Assistant for Hebrew, letter-spacing 0 by default\n- **Signature components** — `PurpleBar`, `YellowUnderline`, `FlightHUD`, `CompassDial`, `CounterUp`, `HeroBg`\n- **Brand assets bundled** — 4 phoenix logos + studio portrait, with a 4-step asset-retrieval fallback chain for any machine\n- **Canonical link set + bio** — auto-included on any footer / about / credits block\n- **Lessons-learned reference** — 15 hardened production patterns (bilingual toggle, mobile hamburger threshold, Anton letter-spacing rules, yellow-span `line-height ≥ 1.0`, IntersectionObserver for GSAP catalogs, GH Pages cert nudge, …)\n- **Pre-flight checklist** — 14 mandatory checks before any visual ships\n\n🔗 **[View yuv-design-system →](skills/yuv-design-system/SKILL.md)**\n\nPairs with `yuv-decks` (cinematic presentation builder), `yuv-viral-video` (short-form video), and `hyperframes` (capture web → MP4). Project brand wins when explicitly specified; otherwise this skill fills the vacuum.\n\n---\n\n## 🆕 NEW: yuv-decks — opinionated cinematic deck builder\n\nBuilds open-slide presentation decks in Yuval Avidani's signature style: 4-act narrative arc (Boarding → Ascent → Cruise → Descent), Yuval voice (plain-language, no jargon, ≤8–12 word bullets, define every name), the flight-themed `JourneyBar` unifying every slide, reusable `UseCase` / `CaseStudy` / `LessonsGrid` templates, and an automatic orchestration of nano-banana cinematic hero images + Hyperframes video moments + Mermaid technical diagrams.\n\nInherits all palette/typography/components from `yuv-design-system` and locks the **Decks mode** (Fly High purple/yellow/grey — the dedicated palette for slides). For talks whose central metaphor IS literal flight, opt into the `cinematic-flight` mode (sky-blue + hot pink) — same arc and JourneyBar, different palette. Never falls through to Neon (web/app palette).\n\n🔗 **[View yuv-decks →](skills/yuv-decks/SKILL.md)**\n\nReference implementation: (the NICE pre-hackathon \"Build Agents That Ship\" deck, May 2026).\n\n---\n\n## 🆕 NEW: Parallax Landing Page Skill\n\nTurn any short video (5–15s) into a **cinematic scroll-driven landing page** where the\nuser's scroll gesture *scrubs the frames in place* — the page itself never scrolls. A single\nlocked viewport with five dramatic text scenes that crossfade in/out as the frame index\nadvances, an Anton + Caveat typography contract, off-white #f5f1ea (never pure white), a\nvignette + film-grain overlay, a loader that preloads every frame, and a final scene that\nholds the CTAs (no separate end section). Built on the proven `parasites/` reference repo\n(GitHub Desktop / Marcus the white lion / Hope the cheetah landings).\n\n🔗 **[View the Parallax Landing Page Skill →](skills/parallax-landing-page/SKILL.md)**\n\n**What's inside:**\n- `scripts/extract_frames.py` — ffprobe + ffmpeg `-q:v 2` extraction with auto-computed `scrollBudget` (≈26 px per frame, clamped to `[2500, 8000]`)\n- `assets/parallax.js` — virtual-scroll controller: wheel/touch/keyboard intercepted, document never moves, RAF-driven lerp (0.22) for buttery frame scrubbing\n- `assets/style.css` — locked-body landing page CSS (`body.scrub-page`), 5-scene grid, vignette + grain, Anton/Caveat/Inter font stack\n- `assets/landing-page-template.html` — full template with `{{placeholders}}` for slug, scenes, CTAs\n- `assets/hub-template.html` — 3-up showcase grid template for multi-landing projects\n- `references/copy-guide.md` — the five-beat arc (Hook → Origin → Stakes → Moment → Resolution) with three real cinematic examples\n- `references/showcase-integration.md` — exact edits to grow an existing hub (card + nav + chain CTA)\n\n**One-shot:** point the script at a video, capture the JSON, fill in the template, serve with `python -m http.server`.\n\n```bash\npython skills/parallax-landing-page/scripts/extract_frames.py demo.mp4 ./out/demo\n```\n\n🎬 **See it live:** the three landing pages built with this skill ship in\n[`examples/parasites/`](examples/parasites/) — GitHub Desktop, Marcus the white lion,\nand Hope the cheetah. After cloning:\n\n```bash\ncd examples/parasites && python -m http.server 8000\n# open http://localhost:8000/\n```\n\n> Note: distinct from `video-to-landing-page` (which uses an evenly-spaced\n> scroll-listener approach). This skill is the \"Apple AirPods Pro\" variant — every frame,\n> locked body, dramatic typography, narrative beats.\n\n---\n\n## 🆕 NEW: Video-Edit Skill (with Interactive Transcript Editor)\n\nA complete captioned-video pipeline that **pauses for human transcript approval** before the\nlong final render — the support mechanism that makes captions perfect (especially Hebrew).\nTranscribes any video with `faster-whisper` (defaults to `large-v3`), applies a configurable\ncorrections dictionary, opens an **interactive browser editor** with video preview, per-line\ninline editing, dictionary apply, find/replace and **optional in-browser LLM suggestions via\nWebLLM** (Qwen2.5-3B / Llama-3.2-3B over WebGPU — runs entirely offline once cached). On\napproval, redistributes word timings and generates a HyperFrames composition with\nliquid-glass caption pills, drifting blob background, liquid morph wipes, and optional\nbehind-subject text via background removal. Supports English, Hebrew and any\nWhisper-supported language.\n\n🔗 **[View the Video-Edit Skill →](skills/video-edit/SKILL.md)**\n\n**What's inside:**\n- `transcribe.py` — faster-whisper CPU/CUDA with safe fallback\n- `make_review.py` + `apply_review.py` — emit & ingest a `transcript_review.txt` round-trip\n- `transcript-editor/index.html` — self-contained webapp (RTL-aware, autosaves to localStorage)\n- `gen_body.py` — generator for editorial + matrix liquid-glass caption sub-composition\n- `host-template.html` + `liquid-blobs.html` + EN/HE parallax-outro templates\n- `corrections-hebrew.md` — curated dictionary of common Hebrew Whisper mishears\n\n---\n\n## 🆕 NEW: Video-to-Landing-Page Skill\n\nDrop a video, get a cinematic **scroll-driven landing page** — Apple-style sticky hero where\nscrolling progresses the visible frame through the video. Extracts evenly-spaced frames via\nffmpeg, builds a self-contained `index.html` with a `requestAnimationFrame`-driven scroll\nlistener, and includes headline / sections / CTA below. Respects the `yuv-design-system` typography\n(Anton + Inter for English, Rubik + Assistant for Hebrew) and Fly High purple palette by default.\n\n🔗 **[View the Video-to-Landing-Page Skill →](skills/video-to-landing-page/SKILL.md)**\n\n**One command from video to deployable page:**\n\n```bash\npython skills/video-to-landing-page/references/extract-frames.py demo.mp4 ./landing-demo --build-html\n```\n\nThen customise `__HEADLINE__`, `__TAGLINE__`, `__CTA_TEXT__` in the generated `index.html`\nand drop the folder on Vercel / Netlify / Cloudflare Pages.\n\n---\n\n## 🚀 NEW: Yuv-Viral-Video Skill\n\n**Just added!** The complete signature-style video editor that turns raw selfie or screen-share footage into a viral short-form video the way YUV.AI ships them — Apple-style liquid-glass cards with real backdrop blur, dark-mode polish, MrBeast-paced cuts, video-title karaoke captions, content-relevant motion graphics, no fake content, **never covering the speaker's face**, always rendering BOTH 9:16 and 16:9, always saving with `_V` backup suffix. Hebrew is rendered in Rubik Black (with proper RTL via `python-bidi`), English in Anton uppercase. Encodes every painful lesson from real edits as a hard rule the next render automatically respects.\n\n🔗 **[View the Yuv-Viral-Video Skill →](skills/yuv-viral-video/SKILL.md)**\n\n**What it does:**\n- Apple-style frosted glass cards: rounded corners + `gblur sigma=24` backdrop + `alphamerge` masks so the blur stops cleanly at the rounded shape (no rectangular leak past the corners)\n- Auto-routes the layout per archetype: screen-share footage shows the screen content (top 2/3) with rounded speaker PIP at bottom 1/3; full-frame selfie shows the speaker prominently with cards in the safe top-banner zone (face never covered)\n- Per-segment dynamic motion (push-in / pull-out / snap-in / dolly / shake) baked into the extract chain with subtle MrBeast-style amounts\n- Real animated diagrams: bar chart (uses actual category names from transcript, never fake numbers), flow diagram with glowing brand-color nodes connected by animated dashed arrows\n- Word-by-word video-title karaoke: scale-pop entrance with `\\fscx150\\fscy150\\t(0,140,\\fscx100\\fscy100)`, accent words get yellow + rotation jitter, English brand tokens auto-route to Anton uppercase\n- Bottom key-moment glass strips that summarise the current beat in 1-4 words (Hebrew right + English left + yellow accent stripe)\n- ElevenLabs SFX kit (impact, bass_drop, whoosh, riser, ding, glitch, typing) mixed at the right output times via `adelay`\n- Two-pass loudnorm to -14 LUFS / -1 dBTP / LRA 11 (TikTok / Reels / YouTube standard)\n- **Versioning**: every render writes a NEW `final_*_V.mp4` — previous versions are never overwritten\n- **Truth contract**: every word on every card is auditable against the actual transcript. No \"$$$\" placeholders, no \"100% הבנה\" if the speaker didn't say it, no \"לינק בביו\" if no link was promised\n\n---\n\n## 🆕 NEW: Meta Ads Skill\n\n**Just added!** A battle-tested skill for pulling, analyzing, and managing Meta ad performance (Facebook, Instagram, Messenger, Click-to-WhatsApp, Threads) via the Marketing API. No guesswork, no generic dashboards — the scripts auto-discover your accounts, decode status codes and currency units, handle rate limits and async insights jobs, and guard write actions behind explicit confirmations.\n\n🔗 **[View the Meta Ads Skill →](skills/meta-ads/SKILL.md)**\n\n**What it does:**\n- Pulls ad-level, ad-set-level, campaign-level, and account-level insights with full breakdowns (platform, placement, age, gender, country, device)\n- Auto-discovers every ad account your token can see, ranked by last-30-day spend — so you never paste the wrong `act_` ID\n- Handles both auth paths out of the box: System User tokens (never expire, BM-owned accounts) and long-lived user tokens (60-day, sees personal Instagram-boost accounts)\n- Ships a token-exchange script for the short→60-day swap — no more fumbling through Graph Explorer docs\n- Creative fatigue detection, anomaly detection vs. prior period, CTWA-aware analysis playbooks\n- Write actions (pause, budget update, duplicate) gated behind explicit confirmation with impact summaries and rollback patterns\n- Baked-in troubleshooting for every failure mode we hit the hard way: sandbox proxy blocks, 37-month data wall, DELETED-status error 1815001, minor-unit currency bugs, IG boosts invisible to SU tokens, and more\n\n---\n\n## 🆕 NEW: Google Workspace CLI Skill\n\n**Just added!** A comprehensive skill that teaches AI agents to use the `gws` CLI — one command-line tool for **all** of Google Workspace: Drive, Gmail, Calendar, Sheets, Docs, Slides, Chat, Tasks, Admin, Meet, Forms, Keep, and more. Your agent can now manage your entire Google Workspace without custom tooling.\n\n🔗 **[View the Google Workspace CLI Skill →](skills/google-workspace-cli/SKILL.md)**\n\n**What it does:**\n- Enables AI agents to interact with every Google Workspace API via `gws` CLI commands\n- Covers Drive, Gmail, Calendar, Sheets, Docs, Slides, Chat, Tasks, Meet, Forms, Admin, Keep, Apps Script, and more\n- Includes 100+ ready-to-use command examples with real JSON payloads\n- MCP server integration for Claude Desktop, VS Code, Gemini CLI, and other MCP clients\n- Multiple auth workflows: interactive, headless/CI, service accounts, and multi-account support\n- Built on Google's Discovery Service — automatically picks up new API endpoints\n\n---\n\n## 🧠 Honest Agent Skill\n\nA universal skill that configures ALL your AI coding agents to be honest, objective, and non-sycophantic. Run it once, and every agent (Claude Code, GitHub Copilot, Cursor, Windsurf, etc.) will stop telling you what you want to hear and start giving you honest, direct feedback.\n\n🔗 **[View the Honest Agent Skill →](skills/honest-agent/SKILL.md)**\n\n**What it does:**\n- Configures honest, objective communication across 7 verified AI coding agents\n- One-time setup that updates all your agent instruction files\n- Supports both project-level and global configuration\n- Works with Claude Code, Copilot, Cursor, Windsurf, Cline, Aider, and Continue.dev\n\n---\n\n

\n A curated collection of AI agent skills for enhanced coding assistance\n

\n\n

\n Supercharge your AI coding agents with specialized knowledge and production-ready patterns\n

\n\n

\n Skills •\n Quick Start •\n Usage •\n Contributing\n

\n\n---\n\n## 👋 About Yuval Avidani\n\n\n\n\n\n\n
\n\"Yuval\n\n\n### **Yuval Avidani** — *AI Builder & Speaker · Founder of YUV.AI*\n\n**🏆 2× GitHub Star** · **🌟 AWS Gen AI Superstar** · **📺 AI commentator on Channel 12 News (Israel)** · **🎤 Enterprise AI Trainer**\n\n> Founder of **YUV.AI** — leading Hebrew-speaking AI educator and technical innovator. Builds, teaches, and ships AI that actually works. Open-source maintainer of 110+ projects spanning AI agents, video pipelines, design systems, and developer tools.\n\n**Hebrew bio:** *יובל אבידני — מגיש פינת AI בחדשות 12, GitHub Star כפול, AWS Gen AI Superstar, מייסד YUV.AI. בונה, מלמד, ומשגר AI שעובד.*\n\n
\n\n### 🔗 Find me everywhere\n\n

\n \n \n \n \n \n \n \n \n \n

\n\n### 💼 Work with me\n\n- **🎓 Practical Claude Desktop course** → \n- **🚀 Enterprise AI training & speaking** → DM on [LinkedIn](https://www.linkedin.com/in/yuval-avidani-87081474/) or [@yuvalav](https://x.com/yuvalav)\n- **🤝 Built something with this stack?** Tag [@yuvalav](https://x.com/yuvalav) on X or DM [@yuval_770](https://instagram.com/yuval_770) on Instagram — I'll feature the best work\n- **⭐ Star this repo** if it saves you time — helps other builders find it\n\n---\n\n## ✨ What Are Agent Skills?\n\nAgent skills are **specialized knowledge modules** that enhance AI coding agents with domain-specific expertise. They provide:\n\n- 🎯 **Focused Knowledge** — Only what the AI doesn't already know\n- 📝 **Production-Ready Code** — Working examples, not abstract concepts \n- ⚡ **Quick Activation** — Trigger keywords for instant context\n- 🔄 **Reusable Patterns** — Copy once, use everywhere\n\n---\n\n## �️ Understanding Agent Skills (Open Standard)\n\n### What is SKILL.md?\n\n**SKILL.md** is a simple, open format for giving AI agents new capabilities and expertise. It's an open standard that works across multiple AI coding agents and platforms.\n\n> **Agent Skills** are folders of instructions, scripts, and resources that your AI agent can discover and use to perform tasks more accurately and efficiently.\n\n### Why Use Agent Skills?\n\n**For You (Developer)**:\n- 📚 **Learn Faster** — Pre-built knowledge from experts in your domain\n- 🎯 **Better Results** — Agents have context to work more accurately\n- ♻️ **Reuse Everywhere** — Same skill works across Copilot, Claude Code, Cursor, Windsurf, and more\n- 🔐 **Version Control** — Skills are part of your repository, tracked in git\n\n**For Your Team**:\n- 📖 **Capture Knowledge** — Document processes and best practices\n- 🚀 **Share Skills** — Distribute expertise across projects\n- 🏢 **Enterprise Use** — Centralize organizational knowledge\n\n**For AI Agents**:\n- 🧠 **Domain Expertise** — Perform specialized tasks reliably\n- 🔧 **New Capabilities** — Enable new workflows and automations\n- 🎓 **Context** — Access procedures, scripts, and examples on demand\n\n### What Can You Do With Agent Skills?\n\nAgent skills enable AI agents to:\n- **Domain Expertise**: Package specialized knowledge (legal review, data analysis, deployment procedures)\n- **New Capabilities**: Create presentations, build MCP servers, analyze datasets, debug failures\n- **Repeatable Workflows**: Turn multi-step tasks into consistent, auditable processes\n- **Interoperability**: Reuse skills across different compatible agent products\n\n---\n\n## 📚 How Agent Skills Work (Step-by-Step for Beginners)\n\n### Step 1: Understanding the File Structure\n\nEach skill is a simple folder with a `SKILL.md` file inside:\n\n```\nyour-project/\n├── .github/skills/ # GitHub Copilot project skills location\n│ └── my-skill-name/\n│ ├── SKILL.md # Main skill file (required)\n│ ├── script.js # Optional: supporting scripts\n│ └── examples.md # Optional: more examples\n├── .claude/skills/ # Claude Code project skills location\n│ └── my-skill-name/\n│ └── SKILL.md\n└── ~/.copilot/skills/ # Personal skills (available across all projects)\n └── my-skill-name/\n └── SKILL.md\n```\n\n### Step 2: Understanding SKILL.md Format\n\nEvery `SKILL.md` file has two parts:\n\n**YAML Frontmatter** (metadata):\n```yaml\n---\nname: my-skill-name # Unique identifier (lowercase, hyphens for spaces)\ndescription: What this skill # When should the agent use this skill?\nlicense: MIT # Optional: license information\n---\n```\n\n**Markdown Body** (instructions):\nThe rest of the file contains instructions and examples for the AI agent.\n\n### Step 3: A Real Example\n\nHere's what a simple `SKILL.md` looks like:\n\n```markdown\n---\nname: github-actions-debugging\ndescription: Guide for debugging failing GitHub Actions workflows. Use this when asked to debug CI/CD failures or workflow issues.\n---\n\n# Debugging GitHub Actions Workflows\n\nWhen debugging a failing workflow:\n\n1. **Check the job logs** - Look for error messages and stack traces\n2. **Review recent changes** - What changed since the last successful run?\n3. **Test locally** - Reproduce the issue in your local environment\n4. **Fix and validate** - Make changes and verify they work\n\n## Common Issues and Solutions\n\n### Issue: Workflow fails with \"Command not found\"\n**Solution**: Install the required tool in your workflow step\n\n### Issue: Permission denied when running scripts\n**Solution**: Add `chmod +x script.sh` before running the script\n```\n\n### Step 4: Agent Discovers and Uses Your Skill\n\nWhen you ask your AI agent to do something:\n\n```\n\"Debug why our GitHub Actions workflow is failing\"\n```\n\nThe agent will:\n1. ✅ **Recognize** that you're asking about GitHub Actions\n2. ✅ **Find** the `github-actions-debugging` skill in your project\n3. ✅ **Load** the `SKILL.md` file into its context\n4. ✅ **Follow** the instructions to help you debug\n\n### Step 5: Supported Locations\n\n**Project Skills** (specific to one repository):\n- `your-repo/.github/skills/` — For GitHub Copilot\n- `your-repo/.claude/skills/` — For Claude Code\n\n**Personal Skills** (available across all projects on your machine):\n- `~/.copilot/skills/` — For GitHub Copilot (Copilot CLI and VS Code)\n- `~/.claude/skills/` — For Claude Code\n\n**Organization/Enterprise Skills** (coming soon):\n- Enterprise-level skills support is in development\n\n---\n\n## 🤝 Supported AI Agents\n\nAgent Skills work with these AI coding agents:\n\n| Agent | Support | Location |\n|:------|:--------|:---------|\n| **GitHub Copilot** | ✅ Full Support | `.github/skills/` or `~/.copilot/skills/` |\n| **Claude Code** | ✅ Full Support | `.claude/skills/` or `~/.claude/skills/` |\n| **Cursor** | ✅ Full Support | `.cursor/rules/` |\n| **Windsurf** | ✅ Full Support | `.windsurf/rules/` |\n| **VS Code Insiders** | ✅ Full Support | Agent mode with skills |\n| **VS Code (Stable)** | ⏳ Coming Soon | Support coming in future release |\n\n---\n\n## 📦 Available Skills\n\n| Skill | Description | Keywords |\n|:------|:------------|:---------|\n| **[aws-harness](skills/aws-harness/SKILL.md)** 🆕 | Build a new agent AND deploy it easily on AWS, or wrap + ship an agent you already have, via the Bedrock AgentCore harness - CLI (create/dev/deploy/invoke) plus SDK-wrap and FastAPI/Docker/ECR paths, every import verified | `agent harness`, `AWS harness`, `AgentCore`, `agentcore create`, `agentcore deploy`, `deploy agent on AWS`, `bring your own agent` |\n| **[director](skills/director/SKILL.md)** 🆕 | Award-winning director's brain — a 28-chapter film + AI-production Bible (~108k words) wrapped in a gated, interrogation-first workflow that locks the IDEA and SCRIPT before generating a single pixel, then plans shots/camera/light/sound, then generates. Honest neuroscience (no dopamine-as-pleasure myths), current AI image/video model selection, character-consistency SOP, first/last-frame keyframing | `director`, `video`, `film`, `teaser`, `trailer`, `storyboard`, `script`, `screenwriting`, `documentary`, `סרטון`, `טיזר`, `תסריט`, `סטוריבורד` |\n| **[yuv-design-system](skills/yuv-design-system/SKILL.md)** 🆕 | Yuv's canonical visual brand — Fly High purple (default) + Warm Editorial pink, Anton/Inter + Rubik/Assistant, signature components, brand assets, canonical bio/links, 15 hardened production patterns | `design system`, `brand`, `Fly High`, `Warm Editorial`, `Anton`, `Rubik`, `PurpleBar`, `YellowUnderline`, `FlightHUD`, `yuv.ai` |\n| **[yuv-decks](skills/yuv-decks/SKILL.md)** 🆕 | Open-slide cinematic deck builder — 4-act narrative arc, Yuval voice, JourneyBar, reusable templates, automatic nano-banana + Hyperframes orchestration | `deck`, `presentation`, `slides`, `talk`, `keynote`, `open-slide`, `cinematic deck`, `מצגת`, `שקפים` |\n| **[yuv-viral-video](skills/yuv-viral-video/SKILL.md)** | Short-form video editor — liquid-glass cards, dark-mode, MrBeast-paced cuts, karaoke captions, ALWAYS 9:16 + 16:9, never covers the speaker's face | `viral video`, `reel`, `short`, `selfie edit`, `liquid glass`, `karaoke captions`, `ויראלי`, `ריל` |\n| **[parallax-landing-page](skills/parallax-landing-page/SKILL.md)** 🆕 | Cinematic scroll-scrub landing page from a short video — locked body, virtual scroll, 5 crossfading scenes, Anton + Caveat typography | `parallax landing`, `scroll-scrub`, `frame-by-frame`, `Anton`, `Caveat`, `locked hero`, `virtual scroll`, `video to landing page` |\n| **[video-edit](skills/video-edit/SKILL.md)** 🆕 | Captioned-video pipeline with interactive transcript editor + WebLLM suggestions before render | `video edit`, `captions`, `transcribe`, `Hebrew`, `WebLLM`, `liquid-glass captions`, `HyperFrames` |\n| **[video-to-landing-page](skills/video-to-landing-page/SKILL.md)** 🆕 | Apple-style scroll-driven landing page from any video — frames scrub on scroll | `landing page from video`, `scroll-frame`, `Apple scroll`, `scrub on scroll` |\n| **[meta-ads](skills/meta-ads/SKILL.md)** 🆕 | Meta Marketing API — Facebook, Instagram, CTWA ad insights, fatigue analysis, write actions | `Meta ads`, `Facebook ads`, `Instagram ads`, `Marketing API`, `ROAS`, `CPA`, `CTR`, `CTWA`, `creative fatigue` |\n| **[google-workspace-cli](skills/google-workspace-cli/SKILL.md)** 🆕 | Google Workspace CLI (`gws`) — Drive, Gmail, Calendar, Sheets, Docs, Chat & more | `gws`, `Google Workspace`, `Google Drive`, `Gmail`, `Google Calendar`, `Google Sheets`, `MCP` |\n| **[copilot-sdk](skills/copilot-sdk/SKILL.md)** | GitHub Copilot SDK for building agentic applications | `Copilot SDK`, `GitHub SDK`, `agentic app`, `embed Copilot` |\n| **[honest-agent](skills/honest-agent/SKILL.md)** 🆕 | Configure all AI agents for honest, objective feedback | `honest agent`, `no sycophancy`, `objective`, `contradict me` |\n| **[aws-strands](skills/aws-strands/SKILL.md)** | The agent framework (the \"brain\") - write agent logic, tools & multi-agent systems; deploy via aws-harness | `Strands`, `agent framework`, `agents as tools`, `Swarm`, `Graph`, `BedrockModel` |\n| **[aws-account-management](skills/aws-account-management/SKILL.md)** | AWS Organizations, IAM, billing & multi-account | `AWS Organizations`, `IAM`, `SCPs`, `Cost Explorer` |\n| **[langchain](skills/langchain/SKILL.md)** | LangChain/LangGraph pipelines and agent workflows | `LangChain`, `LangGraph`, `RAG`, `LCEL` |\n| **[vercel](skills/vercel/SKILL.md)** | Vercel deployment, serverless, and edge functions | `Vercel`, `serverless`, `edge function` |\n| **[railway](skills/railway/SKILL.md)** | Railway platform deployment and configuration | `Railway`, `deploy container` |\n| **[cloudflare](skills/cloudflare/SKILL.md)** | Cloudflare Workers, Pages, D1, R2, KV & AI | `Cloudflare`, `Workers`, `D1`, `R2`, `edge computing` |\n| **[figma](skills/figma/SKILL.md)** | Figma API, component code generation & design tokens | `Figma API`, `design tokens`, `Figma to code` |\n| **[fal-ai](skills/fal-ai/SKILL.md)** | Serverless AI image/video generation with fal.ai | `fal.ai`, `Flux`, `SDXL`, `AI image generation` |\n| **[mongodb](skills/mongodb/SKILL.md)** | MongoDB & Mongoose queries, aggregation pipelines | `MongoDB`, `Mongoose`, `aggregation`, `NoSQL` |\n| **[bun](skills/bun/SKILL.md)** | Bun JavaScript runtime, bundler & test runner | `Bun`, `bun.sh`, `JavaScript runtime` |\n| **[owasp-security](skills/owasp-security/SKILL.md)** | OWASP Top 10 security vulnerabilities & prevention | `OWASP`, `security`, `XSS`, `SQL injection`, `CSRF` |\n| **[shabbat-times](skills/shabbat-times/SKILL.md)** | Jewish calendar data and Shabbat times integration | `Shabbat times`, `Hebcal`, `Zmanim` |\n| **[copilot-docs](skills/copilot-docs/SKILL.md)** | GitHub Copilot custom instructions reference | `copilot-instructions.md` |\n| **[nano-banana-pro](skills/nano-banana-pro/SKILL.md)** | Google Gemini 3 Pro Image generation | `Nano Banana Pro`, `Gemini 3 Pro Image` |\n| **[github-trending](skills/github-trending/SKILL.md)** | GitHub trending repositories scraping | `GitHub trending`, `trending repos` |\n| **[ux-design-systems](skills/ux-design-systems/SKILL.md)** | Design systems and component libraries | `design system`, `design tokens`, `theming` |\n| **[web-accessibility](skills/web-accessibility/SKILL.md)** | WCAG compliance and accessibility patterns | `accessibility`, `a11y`, `WCAG`, `ARIA` |\n| **[mobile-responsiveness](skills/mobile-responsiveness/SKILL.md)** | Responsive design and mobile-first patterns | `responsive`, `mobile-first`, `breakpoints` |\n| **[analytics-metrics](skills/analytics-metrics/SKILL.md)** | Data visualization and analytics dashboards | `analytics`, `dashboard`, `charts`, `KPI` |\n| **[mermaid-diagrams](skills/mermaid-diagrams/SKILL.md)** | Mermaid diagram syntax for visualizations | `Mermaid`, `flowchart`, `sequence diagram` |\n| **[local-llm-router](skills/local-llm-router/SKILL.md)** | Route queries to local LLMs in air-gapped networks with Serena MCP | `local LLM`, `Ollama`, `LM Studio`, `air-gapped`, `Serena`, `model routing` |\n| **[x-twitter-scraper](skills/x-twitter-scraper/SKILL.md)** | X/Twitter data extraction — tweet search, user lookup, followers, media, monitoring | `Twitter`, `X`, `scraper`, `OSINT`, `tweets`, `followers` |\n\n---\n\n## 🚀 Quick Start\n\n### 1. Clone the Repository\n\n```bash\ngit clone https://github.com/yuval-avidani/ai-agents-skills.git\n```\n\n### 2. Copy Skills to Your Project\n\nChoose your AI coding agent and follow the instructions below.\n\n---\n\n## 💻 Setup Guide: Using Skills in Your Agent (For Beginners)\n\nThis guide walks you through setting up Agent Skills in your favorite coding agent, step by step.\n\n### 🎯 Prerequisites\n\nBefore you start, you need:\n- A coding agent installed (Copilot, Claude Code, Cursor, or Windsurf)\n- A project folder on your computer\n- Basic familiarity with command line or your agent's UI\n\n### 📍 Step-by-Step Setup\n\n#### **Option 1: GitHub Copilot** (Most Popular)\n\n
\n🟢 Setup GitHub Copilot with Skills\n\n**For Project Skills** (specific to one repository):\n\n1. **In your project folder**, create the skills directory:\n ```bash\n mkdir -p .github/skills\n ```\n\n2. **Copy the skills you want** from this repository:\n ```bash\n # Copy a single skill\n cp -r skills/vercel .github/skills/\n \n # Or copy all skills\n cp -r skills/* .github/skills/\n ```\n\n3. **Or manually create a skill**:\n ```\n your-project/\n └── .github/\n └── skills/\n └── my-custom-skill/\n └── SKILL.md\n ```\n\n4. **Open your project in VS Code** and start using Copilot Agent\n5. **Ask Copilot a question** related to your skill, and it will automatically load it!\n\n**For Personal Skills** (available across all your projects):\n\n1. **Find your home directory** (`~` or `C:\\Users\\YourUsername`)\n\n2. **Create personal skills folder**:\n ```bash\n mkdir -p ~/.copilot/skills\n ```\n\n3. **Copy skills there**:\n ```bash\n cp -r skills/vercel ~/.copilot/skills/\n ```\n\n4. **Now all your projects** can use these skills automatically!\n\n**Verify it's working:**\n- Open Copilot Agent\n- Ask about something covered in your skill (e.g., \"How do I deploy to Vercel?\")\n- Copilot will use the skill to help you\n\n
\n\n---\n\n#### **Option 2: Claude Code**\n\n
\n🟣 Setup Claude Code with Skills\n\n**For Project Skills**:\n\n1. **In your project folder**, create the skills directory:\n ```bash\n mkdir -p .claude/skills\n ```\n\n2. **Copy the skills you want**:\n ```bash\n # Single skill\n cp -r skills/langchain .claude/skills/\n \n # All skills\n cp -r skills/* .claude/skills/\n ```\n\n3. **Open your project in Claude Code**\n4. **Chat with Claude** about tasks covered in your skills - it will automatically use them!\n\n**For Personal Skills** (available everywhere):\n\n1. **Create personal skills folder**:\n ```bash\n mkdir -p ~/.claude/skills\n ```\n\n2. **Copy skills there**:\n ```bash\n cp -r skills/aws-harness ~/.claude/skills/\n ```\n\n**Verify it's working:**\n- Open Claude Code in your project\n- Ask about something in your skill\n- Claude will reference and use the skill\n\n
\n\n---\n\n#### **Option 3: Cursor**\n\n
\n🔵 Setup Cursor with Skills\n\n**For Project Rules**:\n\n1. **In your project folder**:\n ```bash\n mkdir -p .cursor/rules\n ```\n\n2. **Copy SKILL.md files** (rename them as rules):\n ```bash\n # Copy a skill as a rule file\n cp skills/figma/SKILL.md .cursor/rules/figma.md\n \n # Or copy multiple\n cp skills/*/SKILL.md .cursor/rules/\n ```\n\n3. **In Cursor Settings**, configure which rules to use:\n - Settings → Rules → Add project rules\n - Point to `.cursor/rules/` folder\n\n4. **Start using Cursor** - it will apply these rules to your context automatically\n\n
\n\n---\n\n#### **Option 4: Windsurf**\n\n
\n🟡 Setup Windsurf with Skills\n\n**For Project Rules**:\n\n1. **In your project folder**:\n ```bash\n mkdir -p .windsurf/rules\n ```\n\n2. **Copy skills as rules**:\n ```bash\n # Copy specific skills\n cp skills/vercel/SKILL.md .windsurf/rules/vercel.md\n \n # Or copy all skills\n cp skills/*/SKILL.md .windsurf/rules/\n ```\n\n3. **Windsurf automatically** discovers rules in `.windsurf/rules/`\n4. **Start building** - Windsurf will use these rules contextually\n\n
\n\n---\n\n### 🎓 What Happens After Setup?\n\nOnce you've set up your skills:\n\n1. **Agent Detects Skills**: Your AI agent scans the skill directories\n2. **Agent Reads SKILL.md**: It reads the name and description from frontmatter\n3. **Agent Activates on Relevance**: When you ask a question matching the description, the agent loads the skill\n4. **Agent Follows Instructions**: Your agent now has the context to help you accurately\n\n### 💡 Example: Using a Vercel Skill\n\n**You have this SKILL.md**:\n```yaml\n---\nname: vercel-deployment\ndescription: Deploying applications to Vercel. Use this when asked about deploying, hosting, or managing Vercel projects.\n---\n```\n\n**You ask your agent**:\n> \"Help me deploy my React app to Vercel\"\n\n**Agent automatically**:\n- ✅ Finds `vercel-deployment` skill\n- ✅ Loads SKILL.md into context\n- ✅ Follows the deployment instructions\n- ✅ Helps you deploy successfully!\n\n### 🐛 Troubleshooting\n\n| Problem | Solution |\n|:--------|:---------|\n| Agent not using skill | Restart your agent, or make sure folder path is correct |\n| Skill file not found | Verify `SKILL.md` is in the right folder and named exactly \"SKILL.md\" |\n| Agent using wrong skill | Make sure skill descriptions are descriptive enough to match your request |\n\n---\n\n## 💻 Usage (Advanced)\n\n---\n\n## 📐 Skill Format\n\nEach skill follows a consistent structure:\n\n```markdown\n---\nname: skill-name\ndescription: Brief description with trigger keywords\n---\n\n# Skill Title\n\nQuick start and core patterns...\n```\n\n### 🎯 Key Principles\n\n| Principle | Description |\n|:----------|:------------|\n| **Concise is key** | Only include what the AI doesn't already know |\n| **Progressive disclosure** | Start with quick start, then advanced patterns |\n| **Concrete examples** | Working code over abstract descriptions |\n| **Trigger keywords** | Include words that activate the skill |\n\n---\n\n## 📁 Repository Structure\n\n```\nai-agents-skills/\n├── 📄 README.md\n├── 📁 skills/\n│ ├── 📁 meta-ads/ # 🆕 NEW!\n│ │ ├── 📄 SKILL.md\n│ │ ├── 📄 requirements.txt\n│ │ ├── 📁 assets/\n│ │ │ └── 📄 env.template\n│ │ ├── 📁 references/\n│ │ │ ├── 📄 setup.md\n│ │ │ ├── 📄 insights-fields.md\n│ │ │ ├── 📄 analysis-playbooks.md\n│ │ │ ├── 📄 write-actions.md\n│ │ │ └── 📄 troubleshooting.md\n│ │ └── 📁 scripts/\n│ │ ├── 📄 meta_client.py\n│ │ ├── 📄 auth_check.py\n│ │ ├── 📄 exchange_token.py\n│ │ ├── 📄 list_accounts.py\n│ │ ├── 📄 list_campaigns.py\n│ │ ├── 📄 fetch_insights.py\n│ │ ├── 📄 creative_fatigue.py\n│ │ ├── 📄 anomaly_detect.py\n│ │ ├── 📄 pause_ad.py\n│ │ ├── 📄 update_budget.py\n│ │ └── 📄 duplicate_ad.py\n│ ├── 📁 google-workspace-cli/ # 🆕 NEW!\n│ │ └── 📄 SKILL.md\n│ ├── 📁 copilot-sdk/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 honest-agent/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 aws-harness/ # 🆕 NEW!\n│ │ └── 📄 SKILL.md\n│ ├── 📁 aws-strands/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 aws-account-management/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 langchain/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 vercel/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 railway/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 cloudflare/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 figma/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 fal-ai/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 mongodb/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 bun/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 owasp-security/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 shabbat-times/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 copilot-docs/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 nano-banana-pro/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 github-trending/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 ux-design-systems/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 web-accessibility/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 mobile-responsiveness/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 analytics-metrics/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 mermaid-diagrams/\n│ │ └── 📄 SKILL.md\n│ ├── 📁 local-llm-router/\n│ │ ├── 📄 SKILL.md\n│ │ └── 📁 references/\n│ │ └── 📄 model-matrix.md\n│ └── 📁 x-twitter-scraper/\n│ └── 📄 SKILL.md\n└── 📁 templates/\n └── 📁 skill-template/\n └── 📄 SKILL.md\n```\n\n---\n\n## 🤝 Contributing\n\nWe welcome contributions! Here's how to add a new skill:\n\n1. **Fork** this repository\n2. **Create** a new skill in `skills/your-skill-name/SKILL.md`\n3. **Follow** the [skill template](templates/skill-template/SKILL.md)\n4. **Submit** a pull request\n\n### Contribution Guidelines\n\n- ✅ Include practical, production-ready code examples\n- ✅ Add trigger keywords in the description\n- ✅ Test all code snippets before submitting\n- ✅ Keep explanations concise but complete\n- ❌ Don't include basic concepts the AI already knows\n- ❌ Don't use placeholder code or TODOs\n\n---\n\n## 🌟 Star History\n\nIf you find this repository useful, please consider giving it a ⭐!\n\n---\n\n## 📄 License\n\nMIT License — Feel free to use, modify, and distribute.\n\n---\n\n

\n \n \n \n \n \n \n \n \n \n

\n\n

\n Made with ❤️ by Yuval Avidani · YUV.AI
\n Building AI agents that ship · Let's Fly High 🚀\n

\n\n

\n \n \"Star\n \n \n \"Follow\n \n

\n", "bytes": 49300, "sha": "74d847132f021361418d82e65ff71f75299908a0b491dc3f5d4e0f345aa1b373", "repo_slug": "hoodini/ai-agents-skills", "fonte": "repo", "truncated": false, "api": "https://agentalog.com/api/listings/skl_hoodini_ai_agents_skills_github_trending_1c469d5d/readme" }