Data sourcing & citations

Vector's promise: public data, every source cited. No black-box scraping, no mystery data.

Where data comes from

1. Public APIs — open, keyless endpoints (GitHub, Wikipedia, weather, geocoders, RSS, …).

2. A page's own markup — its HTML, OpenGraph tags, JSON-LD, headings, links.

3. Open records — census jurisdiction, public geographic data.

Every response cites its sources

The envelope includes a sources array with the exact URLs used:

"sources": ["https://api.github.com/repos/torvalds/linux"]

That means you can verify, attribute, and audit every field you get back.

What Vector will not do

• No harvesting of private personal data.
• No bypassing paywalls or auth.
• No ignoring robots.txt.

This is the line that keeps the platform legitimate and sustainable. See Acceptable use.