About
Built because recalls matter — and manual monitoring doesn't scale.
RecallStream exists because the US government has not offered a unified cross-agency recall API for years, leaving compliance teams to monitor dozens of incompatible agency feeds across 22+ countries manually. We built the aggregation layer they were missing.
Mission
The problem we're solving
openFDA, CPSC, NHTSA, EU Safety Gate, France RappelConso, Health Canada, and UK OPSS each publish recall notices on their own schedule, in their own format, with no unified alert mechanism. Teams resort to bookmark folders, RSS readers, and someone checking a website on Monday morning.
One missed Class I recall can mean regulatory penalties, customer harm, and a reputational crisis that takes years to recover from. The consequence is not hypothetical — it's a weekly occurrence across industries.
RecallStream watches 31 live agency feeds around the clock, normalizes every record to a single schema, and pushes the ones that match your brands, suppliers, or keywords straight to your systems via webhook. You monitor once; we do the work.
By the numbers
How we're different
Aggregated from official agency APIs and open-data feeds.
openFDA, CPSC, NHTSA, EU Safety Gate, Health Canada, UK OPSS and 25 more across 22+ countries.
Free plan, no credit card required.
No credit card required
Data sources
Where the data comes from
All live sources are public-domain, open-government, or CC-licensed government data — no scraping, no intermediaries.
U.S. Food & Drug Administration
Food, drug, and medical device recalls. Published via the FDA's official REST API. Public domain.
Consumer Product Safety Commission
Consumer product hazards, recalls, and corrective actions. Public domain.
National Highway Traffic Safety Administration
Vehicle recalls, defect investigations, and VIN lookup. Public domain.
European Commission Safety Gate
EU non-food consumer product hazard alerts (RAPEX). Licensed CC BY 4.0 — attribution required.
Architecture
How it's built
RecallStream is a NestJS + Prisma + PostgreSQL + BullMQ service running on EU-hosted infrastructure. Agency data is fetched on short poll intervals, deduplicated by source record ID, and stored in a normalized schema. Webhook delivery uses signed HMAC-SHA256 payloads with exponential backoff retry.
Data & licensing
The recall data is public domain
The underlying government recall data is public-domain or openly-licensed — each feed's licence and required attribution are listed on our Sources page. RecallStream's value is the aggregation, normalization, near-real-time delivery, and watchlist matching infrastructure — not the raw data. You can access the raw agency feeds directly; we make it so you don't have to.
Ready to stop missing recalls?
Start free in under 60 seconds — no credit card required.
Have a question first? Reach us at support@recallstream.com