50+ public sources: how Luna monitors competitors without paid APIs

Most competitive intelligence tools depend on expensive APIs or proprietary databases. Luna takes a radically different approach.

Why public sources are enough

Companies leave traces everywhere: press releases, product updates, job postings, patent filings, customer reviews, GitHub commits. Individually, these signals are noise. Cross-referenced by AI, they reveal a strategy.

Our 8 source categories

1. News and media (12 sources)

Google News, TechCrunch, industry-specific press. Captures official announcements and media coverage.

2. Technical community (8 sources)

Reddit, Hacker News, Dev.to, Stack Overflow. Reveals developer sentiment and competitors' technical issues.

3. Product and releases (7 sources)

Product Hunt, GitHub Releases, npm, PyPI, official blogs. Detects feature launches before press releases.

4. Jobs and organization (6 sources)

LinkedIn, Indeed, Greenhouse, Lever. Job postings reveal future strategic priorities.

5. Patents and regulation (5 sources)

USPTO, EPO, SEC EDGAR, EUR-Lex, CNIL. Long-term signals about innovation and compliance.

6. Reputation and reviews (5 sources)

G2, Capterra, Trustpilot, Apple Podcasts, YouTube. The voice of the market and end users.

7. Social and blogs (4 sources)

Mastodon, Bluesky, RSS blogs, podcasts. Informal communication and editorial leadership.

8. Science and research (3 sources)

arXiv, PubMed, HAL. Fundamental R&D and academic publications.

The magic of cross-referencing

A single isolated signal is worthless. But when Luna simultaneously detects an ML job posting, a new GitHub repository, and a filed patent, it understands that a competitor is preparing an AI pivot — before the official announcement.