Fec web workers
Use when moving expensive browser work off the main thread with Web Workers, SharedWorker, worker pools, Comlink, transferable objects, Vite/Webpack worker integration, or UI responsiveness fixes. Do not use for lightweight synchronous work or DOM manipulation; Chinese triggers include Web Worker, background thread, main thread blocking.From its SKILL.md
npx -y skills add bovinphang/frontend-craft --skill fec-web-workersAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 21 stars21 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.4 KB, 400 tokens by cl100k_base, as published. Nobody here has run it
Web Workers
Purpose
Move expensive calculations off the main thread and keep input, scrolling, and animations responsive.
Procedure
- Confirm the bottleneck first: consider workers only if the task exceeds about 10ms, user input delay is obvious, FPS drops, or large file parsing/image processing/encryption calculation.
- Choose a communication model: use
postMessagefor simple tasks, use Comlink for functional RPC, use Worker pool for a large number of parallel small tasks, and use SharedWorker for cross-Tab synchronization. - Design the message protocol: request/response types, error types, cancellation or timeout strategies must be clear.
- Prioritize Transferable Objects when transferring large objects; don’t let structured cloning swallow up the benefits brought by Worker.
terminate()when a component or page is unloaded and verify main thread improvements through Profiler, Performance panel or FPS metrics.
Detailed reference
- Load references/basic-worker.md when you need native Worker files, build tool imports, React lifecycle encapsulation and cleanup examples.
- Load references/advanced-workers.md when Transferable Objects, Comlink, Worker pool, SharedWorker, CSP and SharedArrayBuffer considerations are required.
Constraints
- Cannot access DOM,
window,document,alert()or synchronizelocalStoragewithin Worker. - High frequency communication may be slower than main thread calculations; measure first then split.
- Worker files must be loaded from the same source and are subject to
worker-srcCSP restrictions. - Large binary data is transferred first, and the original reference will become invalid after the transfer.
- Each Worker has an independent JS heap and must actively manage life cycle and memory.
Expected Output
Computationally intensive operations are performed on a background thread, main thread input and scrolling resume smoothly, error and cancellation paths are clear, and workers are terminated on unloading without memory leaks.
What ships with it: 2 files
3.2 KB alongside SKILL.md
references/
- advanced-workers.md1.5 KB
- basic-worker.md1.7 KB