FactCatBot

If you found this page from a User-Agent string in your logs, this is the right place. FactCatBot is the fetcher behind FactCat, a small project that turns verified Wikipedia facts into short documentary videos.

Current user agent:

FactCatBot/1.0 (https://www.factcat.fyi/bot; [email protected]) python-requests/<version>

What it does

• Reads article metadata and image/video file information through the MediaWiki Action API (en.wikipedia.org, commons.wikimedia.org, www.wikidata.org).
• Downloads openly licensed media (public domain / CC0 / CC BY) from upload.wikimedia.org for use in videos, with per-file attribution recorded and credited.
• Occasionally queries other open collections the same way: the Internet Archive (archive.org), the Library of Congress (loc.gov), and NASA's image API.

How it behaves

• Requests to Wikimedia projects are rate-limited process-wide to at most 200 requests/minute with at most 3 concurrent connections, per the Wikimedia API rate-limit guidance.
• 429 responses are honoured: the bot waits the Retry-After interval and backs off exponentially, giving up after 3 retries.
• Results are cached locally, so repeat runs do not re-fetch media.
• Typical traffic is small and bursty: a few dozen requests around each video's production, a handful of videos per week.

Contact

Seeing unwanted traffic, or have any question about the bot? Email [email protected] and we will fix it promptly.