Product Recording
Prepare Remotion Studio for a user-recorded product demo with Canvas Capture, then import the downloaded take into packages/brand as an editable composition. Use for recording Studio interactions, timeline demos, and product shots.
Get the complete skill folder, including its scripts and reference files.
When to use this skill
Use Product Recording when an AI agent needs a reusable SKILL.md workflow for this job: Prepare Remotion Studio for a user-recorded product demo with Canvas Capture, then import the downloaded take into packages/brand as an editable composition. Use for recording Studio interactions, timeline demos, and product shots.
When not to use it
Skip Product Recording when the task is outside the creativity category, or when a more specific skill in this directory already covers the same workflow with clearer triggers.
How to install
Where will you use it?
mkdir -p ~/.claude/skills/product-recording # Save SKILL.md to ~/.claude/skills/product-recording/SKILL.md
- Claude Code personal install: place the skill and its bundled files at ~/.claude/skills/product-recording/SKILL.md. For a shared project, commit the folder at .claude/skills/product-recording/ instead.
- Current Claude Code detects SKILL.md changes in watched directories during the session. If you just created a top-level skills directory, run /reload-skills. Check that the skill is listed before testing its trigger.
- For Claude chat or Cowork in the Desktop app, manage uploaded skills in Customize > Skills; copying files into a local Claude Code folder is not a Desktop install. Other agents may use different install locations.
Full install guide for Claude, Cursor, and Codex →
Try a starter prompt
After installing the skill, paste this starter prompt into your agent to verify it triggers on a small task.
Use the Product Recording skill for this task: Use Product Recording when this documented workflow matches the task. Confirm any required inputs or files before making changes.
What supports this recommendation?
71/100 provisional content score. This measures documentation, practical detail and source signals. It is not a measured task success rate or a security certification.
- Documentation
- 77/100
- Practical detail
- 68/100
- Source evidence
- 63/100
- Source signals
- 72/100
Model compatibility evidence
- claudeuntested
No model-specific signal or recorded compatibility test was found.
- chatgptuntested
No model-specific signal or recorded compatibility test was found.
- geminiuntested
No model-specific signal or recorded compatibility test was found.
- copilotuntested
No model-specific signal or recorded compatibility test was found.
- llamauntested
No model-specific signal or recorded compatibility test was found.
- perplexityuntested
No model-specific signal or recorded compatibility test was found.
- mistraluntested
No model-specific signal or recorded compatibility test was found.
- grokuntested
No model-specific signal or recorded compatibility test was found.
Inferred compatibility is based on content signals. Check the app, its available tools and setup requirements before running a skill.
What this skill does
# Product recording
Prepare the browser and shot, let the user perform and record the interaction, then bring the original capture into `packages/brand`. This is an internal Remotion workflow; keep this skill in `.agents/skills`.
## Prepare Studio and Chrome
1. Use [canvas-capture-extension](../canvas-capture-extension/SKILL.md) to build and install the extension in its durable directory. Reload the installed extension through `chrome://extensions` after rebuilding; a successful build alone does not refresh Chrome's loaded extension. 2. Use an existing compatible Chrome with Canvas Draw Element enabled. Preserve unrelated browser tabs and unfinished recordings. 3. Find the requested composition's registration and owning package. Follow [studio](../studio/SKILL.md) for dependency/build preparation and server startup, but start in the owning package and open the recording in compatible **Chrome**. Use `bunx remotion studio --no-open` and the exact URL printed by that invocation, including when it reports an existing server. Do not assume port 3000 belongs to this checkout. 4. Open the composition with `studioPixelRatio=4` in the query string, preserving any other URL parameters. For example, if Studio reports port 3001:
```text http://localhost:3001/RollerSkiRoughCut?studioPixelRatio=4 ```
The parameter increases Studio's canvas backing resolution for sharp timeline captures. It does not set the extension's output resolution. `RollerSkiRoughCut` lives in `packages/jonnys-videos`; locate other requested compositions rather than assuming they live in `packages/example`. 5. Verify the requested composition and current timeline are visible. Activate Canvas Capture through its toolbar icon or Extensions menu.
## Prepare an optional shot and hand off
- Arrange the viewport, timeline height/zoom, playhead, and sidebars as needed for the requested demonstration. Do this before framing the shot. - Use the extension's `setup_shot` and `get_shot` WebMCP tools when available. They are exposed after activating the extension. Choose the viewport, one visible element, or a rectangle; inspect a still with `get_shot` and `preview: true`. The crop stays fixed in viewport CSS pixels, so reframe after changing the window or layout. - If those tools are unavailable, use the extension's **Entire page** or **Select area** controls and visually verify the selection. Output zoom and dimensions are separate from `studioPixelRatio`; set them for the intended shot and check the displayed dimensions. - Before handing off, note existing capture files and their modification times in the user's Downloads folder so the new take can be identified later. - Leave the shot ready without starting recording. Tell the user to click **Record**, perform their interaction, stop, and choose **Download recording**. The user controls the cursor and recording start/stop. Pause here until they have recorded; do not substitute an older download or claim an import is done.
## Import the downloaded take
1. After the user finishes, identify the new, fully downloaded capture in `~/Downloads` by filename and modification time relative to the handoff. Exclude incomplete `.crdownload` files. If several takes are plausible, ask which one to use. 2. Keep the original MP4 or WebM with embedded `REMOTION_CAPTURE_DATA` metadata. Do not transcode it before import; the metadata carries editable cursor data. 3. Use [import-canvas-capture](../import-canvas-capture/SKILL.md), targeting `packages/brand` and its composition registration file (currently `src/Root.tsx`). Choose an unused, descriptive composition ID. From `packages/brand`, the local built CLI can be invoked as:
```bash node ../cli/remotion-cli.js experimental-codemod add-canvas-capture \ /absolute/path/to/downloaded-capture.mp4 \ --composition-file src/Root.tsx --id TimelineRecording ```
Follow that skill's dimension, dependency, and conflict guidance. Inspect the capture dimensions and pass `--width` and `--height` to match them when importing the full shot: the default 1920×1080 composition can crop a larger recording because the generated component retains native video dimensions. Preserve the download and existing compositions/assets. 4. Review the generated component, registration, copied public asset, and any dependency changes. Run the relevant brand validation before offloading.
For exports, follow [remotion-render](../remotion-render/SKILL.md) and use Remotion's bundled browser. The Chrome instance used for recording does not need to be passed to the renderer.
## Offload the recording
Finish with [offload-r2](../offload-r2/SKILL.md): upload the imported recording to `remotion.media`, verify that the public object's SHA-256 matches the local file, and replace the composition's local asset reference with the hosted URL. Remove the repository-local media copy only after verification; preserve the original download. Follow that skill's focused validation, then report the composition ID, hosted URL, and changed files. Distinguish a browser ready for recording from a completed, imported, and offloaded take.
Intended uses
- Use Product Recording when this documented workflow matches the task.
Implementation notes
- Review the source instructions and adapt inputs before running the workflow.
Try it, then tell us how it went
A copied prompt does not tell us whether the task worked. Report your result after checking the output. We collect your tool and outcome, never your files or prompt.