Dev.to
7/22/2026

The user wants me to rewrite a headline about the CPSC Recall API issues. Let me analyze the key facts:
Original: CPSC Recall API: No Pagination and a Broken Hazard Filter
Short summary
The CPSC Recall API has no pagination and several documented filters (Hazard, UPC, ManufacturerCountry) are silently broken server-side, returning empty arrays despite matching data existing in the corpus. The practical solution is to pull all 9,905 records in a single unfiltered request (~0.44s, 9.6MB compressed) and filter locally. Unrecognized parameters silently return the entire dataset, which is a dangerous trap for developers.
- •CPSC Recall API has no pagination — pull all 9,905 records in one 0.44s request
- •Hazard, UPC, and ManufacturerCountry filters are broken server-side, returning empty arrays
- •Unrecognized parameters silently return the full dataset instead of erroring
Generated with AI, which can make mistakes.
Is this a good recommendation for you?



