React Scan
Capture and analyze Remotion Studio React render profiles through development-only WebMCP tools. Use when diagnosing unnecessary re-renders, slow commits, frame-reactive component trees, or when comparing render performance before and after a change.
Get the complete skill folder, including its scripts and reference files.
When to use this skill
Use React Scan when an AI agent needs a reusable SKILL.md workflow for this job: Capture and analyze Remotion Studio React render profiles through development-only WebMCP tools. Use when diagnosing unnecessary re-renders, slow commits, frame-reactive component trees, or when comparing render performance before and after a change.
When not to use it
Skip React Scan when the task is outside the analysis 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/react-scan # Save SKILL.md to ~/.claude/skills/react-scan/SKILL.md
- Claude Code personal install: place the skill and its bundled files at ~/.claude/skills/react-scan/SKILL.md. For a shared project, commit the folder at .claude/skills/react-scan/ 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 React Scan skill for this task: Use React Scan 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
# React Scan
Use targeted captures to optimize measured Studio interactions. Do not infer a performance problem from source alone when the workflow can be reproduced.
## Record a capture
Start from the repository root:
```sh bun run react-scan:capture ```
For the Brand Studio, run:
```sh bun run react-scan:capture:brand ```
Wait for Studio to finish building, open the printed Studio URL with the browser, and let the page settle. The React Scan WebMCP tools are only registered in this development capture mode.
Call `start_react_scan_recording`, then use browser-use to perform one named interaction three times. Call `stop_react_scan_recording` immediately afterwards. Keep the recording short so startup, HMR, and unrelated interactions do not dominate the data. Call `get_react_scan_recording` to return the raw events and an agent-oriented summary.
After retrieving the recording, stop Studio with Ctrl+C. Recordings only live in the WebMCP result; React Scan does not send them to a separate HTTP collector.
React Scan replaces the React DevTools profiling-hook channel while active. Do not record with the React DevTools Timeline Profiler at the same time.
## Analyze the evidence
Read the `summary` returned by `get_react_scan_recording` first. Confirm the capture contains commit events and inspect `profilingHooksStatuses`; do not interpret a missing profiling channel as an idle application.
Prioritize components with high total self duration, repeated expensive renders, or a clear user-visible slow commit. Inclusive duration contains child work and can double-count a subtree, so use it to locate an expensive boundary, then use self duration and the raw tree to identify the actual work.
Use render causes as evidence, not automatic prescriptions:
- High `parentRenderCount` with no relevant prop, state, context, or hook change suggests a stable memoization boundary. - Repeated changed props suggest checking referential stability at the source location before adding memoization. - Hook, state, or context changes require following the owning update; wrapping the component in `memo()` will not block them. - Source locations identify JSX call sites. Search for the component definition before editing.
For timeline-position findings, also read `../studio-perf/SKILL.md` before changing subscriptions or memoization boundaries.
## Verify an optimization
Preserve the baseline capture. After changing code, record the same interaction the same number of times with the same viewport and compare a new capture. Require an improvement in the relevant commit/component metrics without a regression in behavior. Do not optimize every re-render: cheap necessary renders are often preferable to added memoization complexity.
Report the measured before/after change and the checks performed to verify behavior.
Intended uses
- Use React Scan 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.