pexoai/pexo-skills

What it solves

It eliminates the manual effort of video production by automating the entire pipeline from a simple text description, image, or URL to a finished, publish-ready video. Unlike basic tools that return a single raw clip, this project handles scripting, storyboarding, multi-shot sequencing, and post-production (music, subtitles, and transitions) in one workflow.

How it works

It provides a set of "skills" compatible with AI coding agents like Claude Code, Cursor, and OpenClaw. When a user provides an input (text, image, URL, script, or audio), the system automatically selects the best AI model from a library of 10+ video models (such as Seedance 2 and Kling 3.0), generates the necessary assets, and assembles them into a final video. It uses a single API key to manage multiple underlying providers.

Who it’s for

• Content Creators: For making TikToks, Instagram Reels, and YouTube Shorts quickly.
• Marketers: For creating product ads and e-commerce content from product photos or store URLs.
• Founders: For producing brand narrative videos, pitch decks, or SaaS explainer videos.
• Developers: Who want to integrate professional video generation capabilities into their AI agent workflows.

Highlights

• End-to-End Automation: Handles everything from conversation and scripting to rendering and exporting.
• Auto Model Selection: Routes each shot to the most appropriate model based on content type.
• Multi-Input Support: Accepts text, images, URLs, scripts, and audio as starting points.
• Full Post-Production: Includes built-in TTS narration, voice cloning, lip sync, background music, and subtitles.
• Agent Integration: Fully compatible with the Agent Skills standard for use in AI-native IDEs and terminals.