16 hand-picked Media AI skills with features, install steps and official links — for Claude, Cursor and ChatGPT.
ChatGPT-native image generation: text to high-quality images with style control, inpainting and multi-image comparison.
Generative art with p5.js: flow fields, particle systems, seeded randomness with interactive parameter tuning — gallery-grade visuals without coding.
An animated GIF generator optimized for chat: auto-managed size and frame rate with built-in validation — one prompt to a Slack-ready GIF.
Remotion's official skill: generate video programmatically with React — product demos, data animations and personalized video at scale. Change code, change the film.
Programmatic batch renders
OpenAI's official Codex skill: generate, remix and manage short video clips via the Sora API — text straight to cinematic footage.
OpenAI's official skill: batch-generate and edit images for your projects via the Image API — illustrations, icons and assets produced inside your dev flow.
Auto project illustration
An open-source agentic video production system that turns your AI coding assistant into a full video studio. Describe what you want in plain language and the agent handles research, scripting, asset generation, editing and final composition. Ships 12 pipelines, 100+ tools and 700+ skill and knowledge files.
12 video production pipelines
Plain-language one-line output
Runs offline without API keys
Works in five AI assistants
An agent skill that turns one topic into a finished Vox-style paper-collage explainer/ad video — script, collage keyframes, motion, voice-over, music and captions, all automated. Runs on the Atlas Cloud API plus local ffmpeg and works with any coding agent.
One-line topic to finished video
Vox paper-collage visual style
Two approval gates keep control
B-roll, A-roll, C-roll inputs
An AI video skill for Claude Code and Codex that crafts cinematic product videos with Remotion. It ships 100+ shot recipe cards, 161 motion previews and a production-ready template, with real page captures, 2.5D camera moves, beat-synced cuts and film-grade SFX.
Cinematic product videos in Remotion
161 motion previews to pick from
Real page capture + 2.5D camera
Beat-synced cuts, film-grade SFX
Turns articles, copy, screenshots, product notes, subtitles, photos, or user videos into Xiaohongshu carousels, Live Photo motion cards, and paired 21:9 + 1:1 WeChat covers. It includes Editorial and Swiss content planning, layouts, and QA.
Turns a roughly five-second voiceover, opinion, or abstract idea into editorial halftone paper-collage B-roll. It requires approval of the visual metaphor, then the color still, before Gemini Omni Flash assembles the final animation and runs QA.
Voiceover-to-metaphor design
Three-stage human approval
Frame-to-video generation and QA
Turns one character, creature, vehicle, weapon, or prop image into a style-consistent transparent action sequence. It locks identity and limb topology, redraws key poses, creates in-betweens, removes backgrounds, normalizes canvases, and packs spritesheets.
Identity and limb-topology locks
Key-pose redraw and in-betweens
Transparent frames, spritesheets, and previews
Structural validation and visual review
An open-source social-content workspace from Zhejiang University and Peking University teams. Its 112 OpenClaw skills connect trend discovery, planning, copy, image/audio/video production, six-platform publishing, and performance review.
112 executable content and media skills
Planning-to-publishing feedback loop
Local image, audio, and video toolchain
Profiles, asset library, and content calendar
Eleven agent skills for AI short drama and motion-comic production, spanning novel analysis, project development, episodic writing, character assets, image prompts, storyboards, video prompts, generation, editing, and final review.
Eleven connected production stages
Scripts, assets, storyboards, and prompts
Generation, editing, and review loop
Per-skill Claude Code and Codex setup
Give text-only agents structured visual evidence through OCR, layout, semantics, and traceable JSON descriptions, so they can analyze screenshots, interfaces, charts, and photos without guessing unseen pixels.
OCR, layout, and semantic evidence
Six model and CLI provider paths
Private-file permissions and redaction
Two video-analysis skills for Claude Code and Codex: video-shots measures cuts, duration, and motion with ffmpeg to build a shot-by-shot breakdown, while video-sync composites the source beside shot data that scrolls and highlights at each cut.
Measured cuts, duration, and motion per shot
Shot size, category, camera, and rhythm labels
Fifteen code gates and offline reports
Synchronized scrolling shot-data video