Skillget

ozor-skills

agent
Guvenlik Denetimi
Uyari
Health Uyari
  • License — License: MIT
  • No description — Repository has no description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 6 GitHub stars
Code Gecti
  • Code scan — Scanned 1 files during light audit, no dangerous patterns found
Permissions Gecti
  • Permissions — No dangerous permissions requested

Bu listing icin henuz AI raporu yok.

README.md

Ozor Skills

Agent skills for Ozor.ai — the AI video generator for product launches and marketing.

These skills extend Claude Code, Codex CLI, and other AI coding assistants with expert knowledge for creating high-quality videos with Ozor. They follow the open SKILL.md standard.

Skills

🎬 ozor-video-best-practices

Expert guide for writing effective Ozor prompts. Includes prompt formulas for every common video type (product launch, explainer, social ad, feature announcement, investor update, tutorial), scene structure principles, format selection, editing workflow, and common mistakes to avoid.

Triggers on: "make a video", "Ozor prompt", "launch video", "explainer video", "video best practices", and more.

📄 ozor-doc-to-video

Converts any document (PDF, PowerPoint, blog post, README, changelog, URL) into a structured, copy-pasteable Ozor prompt. Reads the document, identifies the best video type, structures scenes, and outputs a complete prompt ready to drop into Ozor.ai.

Triggers on: "turn this doc into a video", "video from this PDF", "video from my pitch deck", "convert this to a video", and more.

📋 ozor-changelog-to-video

Transforms changelogs, release notes, and version updates into polished release announcement videos. Parses change categories, prioritizes user-facing features, and structures scenes with version badges, feature highlights, and update CTAs.

Triggers on: "video from my changelog", "release notes video", "announce this release", "version update video", "what's new video", and more.

🌐 ozor-landing-page-to-video

Converts any landing page or website URL into a video prompt. Extracts hero headlines, value propositions, social proof, and CTAs from web content and produces platform-appropriate video prompts that match the brand's visual identity.

Triggers on: "video from my website", "landing page to video", "turn my homepage into a video", "video from this URL", and more.

📱 ozor-social-batch

Generates a coordinated batch of social media video prompts for multiple platforms from a single input. Each platform gets a unique angle — not just a resized copy — respecting platform-specific formats, durations, and audience conventions.

Triggers on: "social media videos", "videos for all platforms", "Instagram and TikTok video", "social campaign", "batch of social videos", and more.

✨ ozor-prompt-optimizer

Analyzes existing Ozor prompts and rewrites them for better output quality. Scores prompts on six dimensions (specificity, structure, constraints, visual direction, audience awareness, CTA), diagnoses issues, and produces an optimized rewrite.

Triggers on: "improve this prompt", "optimize my Ozor prompt", "why did my video look bad", "fix this prompt", "review my Ozor prompt", and more.

MCP-powered skills

The skills above produce prompts. The skills below actually call the Ozor MCP server (mcp__ozor__*) to generate, export, and manage videos end-to-end. Use these when the user wants a finished, shareable video — not just a prompt.

🎥 ozor-generate

End-to-end text-to-video. Takes a short creative brief and runs the full Ozor pipeline (generate_video → wait_for_job → export_video → wait_for_export → get_embed_code), returning a share URL, editor URL, download URL, and <iframe> embed.

Triggers on: "make a video about ...", "generate a launch video", "create a 30s explainer", "render a video", "spin up an ad for ...".

🗂️ ozor-document-video

End-to-end document-to-video. Takes a real file (PDF, PPTX, DOCX, deck, proposal, report), runs analyze_document to build a structured plan, presents the plan for approval, optionally applies update_plan edits, then runs generate_from_plan → export → embed. Always plan-first so slide visuals are preserved.

Triggers on: "turn this PDF into a video", "video from my pitch deck", "make a video from this PowerPoint", "video from my proposal.docx".

🔗 ozor-url-to-video

End-to-end URL-to-video. Hands analyze_document a URL so Ozor extracts hero headlines, brand colors, social proof, and CTA directly from the page, then runs the plan → generate → export → embed flow.

Triggers on: "make a video from https://...", "video from my landing page", "turn this blog post into a video", "video from our homepage".

📚 ozor-video-library

Manage videos that already exist in the user's Ozor account. Lists recent videos, inspects status, fetches share/download/embed URLs, re-exports, sends follow-up edits via send_message, and edits plans via update_plan.

Triggers on: "list my Ozor videos", "what's the status of my last video", "get the embed code for that video", "re-export at 9:16", "send a follow-up edit".

Installation

Claude Code (via plugin marketplace)

# Add this repo as a marketplace source
/plugin marketplace add ozor-ai/ozor-skills

# Install individual skills
/plugin install ozor-video-best-practices@ozor-skills
/plugin install ozor-doc-to-video@ozor-skills
/plugin install ozor-changelog-to-video@ozor-skills
/plugin install ozor-landing-page-to-video@ozor-skills
/plugin install ozor-social-batch@ozor-skills
/plugin install ozor-prompt-optimizer@ozor-skills

# MCP-powered skills (call the Ozor MCP server end-to-end)
/plugin install ozor-generate@ozor-skills
/plugin install ozor-document-video@ozor-skills
/plugin install ozor-url-to-video@ozor-skills
/plugin install ozor-video-library@ozor-skills

Claude Code (manual)

# Clone and copy to your skills directory
git clone https://github.com/Mintii-Labs/ozor-skills.git
cp -r ozor-skills/skills/* ~/.claude/skills/

npx (one-liner)

npx skills add ozor-ai/ozor-skills

Other agents (Codex CLI, Cursor, Gemini CLI)

Copy the skill folders into your agent's skills directory. The SKILL.md format is universal across all agents that support the standard.

How It Works

These are Encoded Preference skills — they don't call APIs or execute code. They teach your AI assistant the best practices, prompt formulas, and workflows for creating great Ozor videos. When triggered, the assistant produces structured, copy-pasteable Ozor prompts instead of giving vague advice.

About Ozor

Ozor.ai turns any text prompt or document into a professional, launch-ready video in seconds. It supports PDFs, PowerPoints, images, Google Docs, and URLs. Videos are rendered as animated React components at 30 FPS and export to MP4 at up to 4K.

License

MIT

Yorumlar (0)

Sonuc bulunamadi