ipollowork-video-compose — Claude Code/Codex skill for video composition
ipollowork-video-compose is a skill for building and editing the visual assets of an active iPolloWork video, from installed recipes to editable compositions and seek-safe motion.
LicenseRef-Proprietary ★ 7,523 updated 2026-10-29
npx skills add Devin-AXIS/iPolloWork --skill ipollowork-video-compose -a claude-codeWhat ipollowork-video-compose does
The skill works on the video already open in iPolloWork. It reads the saved storyboard and the current entry, then builds the picture from installed recipes, or writes an editable composition of its own when the case calls for it. A local change such as a caption or a theme loads only the rules it touches and leaves the rest of the choreography alone.
A first pass or a serious motion rework needs the motion principles and the real implementation of the chosen recipe read before anything is authored. A continuous motion audition comes before the film grows. Tokens, variables, hooks, media and the timeline stay as they are.
Sound already in the project is not sent off for synthesis. Only new speech and effects go out, and the measured timing is reconciled when they come back. The main agent integrates, repairs and delivers the finished video. The license is proprietary (LicenseRef-Proprietary), so whether it goes into a work project is a call for whoever owns that project.
What it helps with
- assemble an episode from installed recipes without hand-building the timeline
- swap a caption or a still without breaking the rest of the animation
- rework the opening motion and test it before the scene expands
- send only a new line of dialogue to synthesis and match it to recorded audio
- retune the theme across an editable composition
Example prompts for the agent
- Assemble the active iPolloWork video from the installed recipe and leave the timeline as it is.
- Change the caption on the opening frame and keep the rest of the motion untouched.
- Run a motion audition for the new scene before the film grows.
How to install ipollowork-video-compose
1Skills CLI (Vercel)
npx skills add Devin-AXIS/iPolloWork --skill ipollowork-video-compose -a claude-code2Manually: put the files in place
~/.claude/skills/ipollowork-video-compose/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/ipollowork-video-compose/
1Skills CLI (Vercel)
npx skills add Devin-AXIS/iPolloWork --skill ipollowork-video-compose -a codex2Manually: put the files in place
.agents/skills/ipollowork-video-compose/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add Devin-AXIS/iPolloWork --skill ipollowork-video-compose -a cursor2Manually: put the files in place
.cursor/skills/ipollowork-video-compose/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install Devin-AXIS/iPolloWork .agents/skills/ipollowork-video-compose2Manually: put the files in place
.github/skills/ipollowork-video-compose/SKILL.md1Skills CLI (Vercel)
npx skills add Devin-AXIS/iPolloWork --skill ipollowork-video-compose -a gemini-cli2Platform CLI
gemini skills install https://github.com/Devin-AXIS/iPolloWork --path .agents/skills/ipollowork-video-composeManually: put the files in place
.devin/skills/ipollowork-video-compose/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about ipollowork-video-compose
How do I install ipollowork-video-compose in Claude Code?
Run npx skills add Devin-AXIS/iPolloWork --skill ipollowork-video-compose -a claude-code. The skill lands at ~/.claude/skills/ipollowork-video-compose/SKILL.md.
What does ipollowork-video-compose do?
It builds and edits the visual assets of the active iPolloWork video: installed recipes, an editable composition, seek-safe motion. Small changes stay local and the rest of the choreography is kept.
Does it re-synthesize audio that is already in the project?
No. Existing audio is not sent for synthesis; only new speech and sound are handed off, and measured timing is reconciled afterwards.