Question 1

How to add VerifyFetch to a Transformers.js project?

Accepted Answer

Install the @verifyfetch/transformers package, use the CLI to generate hashes for your model, and replace the standard pipeline with verifiedPipeline, passing the manifest URL. This enables automatic verification and resumable downloads with minimal code changes.

Question 2

VerifyFetch vs native fetch for large files?

Accepted Answer

VerifyFetch adds resumability, streaming verification, and corruption detection that native fetch lacks. For AI models or critical assets, it prevents browser crashes and wasted bandwidth, but for simple downloads, native fetch is lighter and sufficient.

Question 3

How to generate hashes for custom files?

Accepted Answer

Use the CLI command 'npx verifyfetch sign <files>' to create SRI hashes, or 'npx verifyfetch sign --chunked' for resumable downloads. Then, serve the generated manifest JSON with your app to enable verification.

Question 4

Can VerifyFetch verify all fetches automatically?

Accepted Answer

Yes, by setting up a Service Worker with createVerifyWorker, it can intercept and verify fetches based on file patterns like *.wasm or *.bin, without modifying your application code, as shown in the README example.

Question 5

What happens if a hash doesn't match?

Accepted Answer

VerifyFetch stops the download immediately and throws an error by default, preventing corrupted files from being used. You can configure fallback URLs or warning modes via the onFail option in the API.

Question 6

Does VerifyFetch support TypeScript?

Accepted Answer

Yes, the library is written in TypeScript and includes full type definitions, ensuring good IDE support and type safety, as seen in the npm package and API reference.

verifyfetch

What is verifyfetch?

Overview

Use Cases

Best For

Related Projects

Found a gem we're missing?

Not Ideal For

Pros & Cons

Pros

Cons

Frequently Asked Questions