Tom Awake

Data tools built on official public APIs. No login, no keys, no proxies, nothing to break.

Most scrapers fight an anti-bot system: they need accounts, proxies and constant patching, and they fail the week you need them. Every tool here reads a public, sanctioned API — FDA, SEC, CMS, USAspending, GLEIF, ClinicalTrials.gov, OpenStreetMap, AT Protocol. Nothing to expire, nothing to get banned.

The API is never the product. Reshaping it into the answer someone needs is:

• A trial with 40 sites is 40 places to call, not one row.
• The FDA publishes when a 510(k) arrived and when it was decided, never the gap — the only number you can plan against.
• Query US revenue the obvious way and you lose Apple, Meta and Microsoft: they report under a different concept. 41% coverage becomes 100%.

Tools

• OSM POI Extractor — local businesses by city and category, with phone, email and website. A legal alternative to scraping Google Maps.
• Clinical Trial Sites & Investigators — one row per site, with the investigator's phone and email.
• NPI Provider Locations — US providers per practice location, with specialty and licence.
• FDA 510(k) Clearances — with the real review time in days.
• SEC Financial Screener — metrics joined across thousands of filers, one row per company.
• LEI Entity Resolver — match a company across LEI, national number, BIC and S&P, with ownership.
• US Federal Award Winners — with each winner's address and small-business status.
• Bluesky Audience Exporter — followers, keyword matches and post likers, enriched.

Every claim comes from a real run. Each README states what the tool cannot do — sparse fields, pagination ceilings, data that lags reality. Where output contains personal data it says so, and says that using it for outreach makes you the controller.

Need a public API in a shape nobody serves? Ask.

---