Argus HQ does not generate regulatory findings — we ingest, structure, and republish public FDA data. The table below lists every source stream, how often we refresh it, and what fields we capture from it. Full detail on ingestion mechanics and AI enrichment is on the Methodology page.
| Stream | Source | Update frequency | Fields captured |
|---|---|---|---|
| Warning Letters | fda.gov Warning Letters listing | Scraped daily, 04:00 UTC | Recipient company, issue date, citations, full letter text, source URL |
| Drug recalls | api.fda.gov/drug/enforcement.json | Every cron run, 10-day lookback | Recalling firm, product description, reason for recall, classification, status |
| Drug approvals | api.fda.gov/drug/drugsfda.json + api.fda.gov/drug/label.json | Every cron run | Applicant, application number, product/brand name, approval date, label summary |
| Form 483 observations | FDA public inspection databases | Pending FDA API key — coverage incomplete today | Inspected facility, observation text, inspection date (where available) |
openFDA’s disclaimer
The recall and approval streams above run through openFDA, the FDA’s own public API. openFDA publishes this disclaimer on every response, and we think it’s worth quoting in full rather than paraphrasing:
“Do not rely on openFDA to make decisions regarding medical care. While we make every effort to ensure that data is accurate, you should assume all results are unvalidated. We may limit or otherwise restrict your access to the API in line with our Terms of Service.”
Read openFDA’s full documentation and terms at open.fda.gov. Argus HQ inherits this disclaimer for every recall and approval record derived from openFDA: our republication does not upgrade the underlying data’s validation status.
For a running, dated log of what was added to the database and when, see Updates.