Hoodini AI Agents Skills
Thirty-nine skills: video editing, landing pages from a clip, decks, AWS, Vercel, Cloudflare, Meta Ads and OWASP
- 291 ★ on the repository
- Last commit
Broad access
Why this level
- `install.sh` installs system packages and a global npm package, writes files and symlinks to the home directory and may save an ElevenLabs key to an env file
- The `meta-ads` skill runs Meta campaigns: it sets up the pixel, launches and edits campaigns, and in bot mode every change waits for confirmation
- `x-twitter-scraper` pulls tweets, followers and media, and `google-workspace-cli` manages Gmail, Drive and Calendar
Install
- Works in
Manual install
curl -sSL https://raw.githubusercontent.com/hoodini/ai-agents-skills/master/install.sh | bashInstalls the 16 creative-stack skills and system dependencies. You can download and read the script first: git clone, then bash install.sh.
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- `install.sh` installs system packages and a global npm package, writes files and symlinks to the home directory and may save an ElevenLabs key to an env file
- The `meta-ads` skill runs Meta campaigns: it sets up the pixel, launches and edits campaigns, and in bot mode every change waits for confirmation
- `x-twitter-scraper` pulls tweets, followers and media, and `google-workspace-cli` manages Gmail, Drive and Calendar
This is third-party code. Review the repository files before installing.
What it does
The hoodini/ai-agents-skills repository is a collection of 39 skills in the `skills/` folder for Claude Code, GitHub Copilot, Cursor and Windsurf. The core is the author's creative stack: `yuv-pilot` routes requests to the rest, `yuv-decks` builds presentations, `yuv-viral-video` and `video-edit` edit short clips with subtitles, `parallax-landing-page` and `video-to-landing-page` turn video into a scroll-scrubbed landing page, `director` guides film work with a 28-chapter handbook. The rest are references for platforms and services: AWS (`aws-harness`, `aws-strands`, `aws-account-management`), Vercel, Railway, Cloudflare, MongoDB, Bun, LangChain, Figma, fal.ai, OWASP Top 10 and WCAG. Separate are `meta-ads` for Meta ad accounts, `google-workspace-cli`, `x-twitter-scraper` for pulling data from X, and `honest-agent`, which appends direct, non-sycophantic feedback rules to agents' instruction files. Install with the `install.sh` script, which sets up the 16 creative-stack skills with `ffmpeg`, Node, Python, `faster-whisper` and the `hyperframes` CLI. The other route is copying individual folders by hand into .claude/skills/, .github/skills/ or ~/.copilot/skills/.
Who it is for. Makers of videos, decks and landing pages, and developers on AWS, Vercel and Cloudflare who want ready skills for these platforms without searching many repositories.
Good fit when
- A 5–15 second clip must become a landing page that scrubs by frame on scroll
- A talk needs a deck with a four-act story arc
- A talking-head video should be edited with subtitles and the transcript reviewed before rendering
- An agent flatters you, and it is time to write direct-feedback rules into every instruction file
Not a fit when
- You need a neutral design: `yuv-design-system` and `yuv-decks` carry the author's brand palettes and a Fly High signature
- The environment must stay untouched: `install.sh` installs `ffmpeg`, Node, Python packages and a global CLI and creates symlinks in home directories
- You need subtitles in Russian: the `video-edit` pipeline is described for Hebrew and English
Example request
Build a scroll-scrubbed landing page from the clip ~/Videos/product.mp4Limitations
- Read `install.sh` before running it: it installs system packages and the global npm package `hyperframes`.
- The `yuv-design-system`, `yuv-decks` and `yuv-viral-video` skills are tied to the author's style: brand palette, Anton and Rubik fonts, a signature.
- The script asks for an ElevenLabs key for `video-use` and `yuv-viral-video`; the step can be skipped.
- `meta-ads`, `google-workspace-cli` and the AWS skills need accounts and access in those services.
How to disable. Remove a skill by deleting its folder from .claude/skills/, .github/skills/ or ~/.copilot/skills/. Symlinks created by `install.sh` sit in ~/.copilot/skills and ~/.agents/skills and are deleted by hand. npm uninstall -g `hyperframes` removes the global CLI.
FAQ
What does the `install.sh` script install?
The creative stack of 16 skills along with `ffmpeg`, Node, uv, `faster-whisper`, the `hyperframes` CLI, Python virtual environments and symlinks into ~/.copilot/skills and ~/.agents/skills. It is repeatable and skips what is installed. Safer to clone the repository first and run bash `install.sh`.
How do I install a single skill without the script?
Clone the repository and copy the folder: cp -r skills/langchain .claude/skills/ for Claude Code or cp -r skills/vercel ~/.copilot/skills/ for personal Copilot skills. To copy all at once use cp -r `skills/`* .claude/skills/.
Which agents are supported?
GitHub Copilot (.github/skills/ or ~/.copilot/skills/), Claude Code (.claude/skills/ or ~/.claude/skills/), Cursor (.cursor/rules/), Windsurf (.windsurf/rules/) and VS Code Insiders in agent mode. For regular VS Code the README says Coming Soon.
How does `parallax-landing-page` differ from `video-to-landing-page`?
`parallax-landing-page` is built for a 5–15 second clip and makes a cinematic landing page with five crossfading scenes. `video-to-landing-page` takes a long video with sections and builds an Apple-style landing page with a sticky hero.
What is `honest-agent` for?
It finds the instruction files of installed agents, asks for scope (project or global) and appends direct, honest-feedback rules. Existing files stay untouched; the skill only appends a block at the end.
Discussion
No questions or reviews yet.
Related
SkillsTwenty-eight skills for Claude Code: voiced video, covers, WeChat layout, Excel, PPT, Dify, GitHub Trending
A set of SKILL.md skills for short-video and content production: topic, script, voice, avatar, editing, covers and motion graphics
Twelve skills that make an agent build single-file HTML animations: slides, protocol diagrams, architecture diagrams and demos for screen recording
A Claude Code skill that downloads a YouTube video, splits it into semantic chapters, cuts chosen clips, translates subtitles and burns bilingual ones in
Motion Skills for AI agents
motion-skills
Over 50 open skills for motion graphics, animation and video: kinetic typography, charts, explainers, TikTok and Reels, WebGL, Manim
A video editing skill for Claude Code and Codex: the agent cuts pauses and filler words, grades color, burns subtitles and adds animated overlays
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| Hoodini AI Agents Skills (this card) | no | 291 | |
| video-use | no | 30,591 | a single editing skill from browser-use that this repository's install.sh installs separately; here a whole collection |
| Designer Skills Collection | no | 3,265 | a set of design skills for research and UX; here design is tied to video, decks and cloud platforms |
README in short
The README opens with the creative-stack install command, then describes the yuv skills pyramid, three palettes and new skills for video, landing pages, ads and Google Workspace. Next come a beginner's explainer on Agent Skills, a table of all skills and copy instructions for Copilot, Claude Code, Cursor and Windsurf.
README badge
Original SKILL.md
--- name: honest-agent description: Detect installed AI coding agents and append honest-feedback rules to their instruction files. Scans for config files (CLAUDE.md, copilot-instructions.md, .cursorrules, etc.), adapts output format per agent, and appends directives that disable sycophancy and enable constructive pushback. Use when setting up honest feedback, disabling people-pleasing, or enabling objective criticism. Triggers on honest agent, objective feedback, no sycophancy, honest criticism, contradict me, challenge assumptions, honest mode, brutal honesty. --- # Honest Agent Configuration One-time setup skill: detects your AI coding agents and appends honesty directives to their instruction files. ## Safety Rule: Append Only **Never overwrite existing instruction files.** Always read the file first (if it exists), then append the new configuration to the end. If the file does not exist, create it. This rule applies to every step below. ## Supported Agents | Agent | Project File | Global File | |-------|-------------|-------------| | Claude Code | `.claude/CLAUDE.md` | `~/.claude/CLAUDE.md` | | GitHub Copilot | `.github/copilot-instructions.md` | - | | Cursor | `.cursorrules` | `~/.cursor/rules/` | | Windsurf | `.windsurfrules` | - | | Cline | `.clinerules` | - | | Aider | `CONVENTIONS.md` | `~/.aider.conf.yml` | | Continue.dev | `.continuerules` | `~/.continue/config.json` | ## Workflow ### Step 1: Detect Agents Scan the project root for each file in the table above. Record which files exist and which agents are present. ### Step 2: Ask Scope Prompt the user to choose: - **Project-level** -- current project only - **Global-level** -- all projects (where the agent supports it) - **Both** ### Step 3: Append Configuration For each detected agent, read the existing file, then append the appropriate block below. **Markdown agents** (Claude Code, Copilot, Cline, Continue.dev): ```markdown