OSINT stands for open-source intelligence: finding and analyzing information from public sources such as news, filings, websites, public social posts and domain records. For a developer it usually means checking what your own systems expose, or researching a company or technology from public data, and a good setup pairs a few lookup tools with a monitor like AyeWatch that tells you when a source changes.
Open source, but not the code kind
The "open source" in OSINT has nothing to do with licenses on GitHub. It means the source is open to anyone: no break-ins, no leaked databases, no private access. AyeWatch's glossary entry on OSINT traces the practice from defense intelligence into journalism, corporate security, fraud investigation and competitive analysis.
What counts as a public source
- News articles and press releases
- Government records: court filings, company registries, securities filings
- Company websites: careers pages, team pages, changelogs
- Public social media posts and profiles
- Research papers and patents
- Domain registration data and TLS certificates
- Archived copies of web pages
The value usually comes from combining them. A new job listing, a patent filing and a changed product page mean more together than any one does alone.
A simple OSINT loop
| Step | What you do | Example tools |
|---|---|---|
| Define | Write down the question and its limits | A notes file |
| Collect | Pull the public data | Google operators, crt.sh, SEC EDGAR, theHarvester |
| Verify | Cross-check dates, copies and sources | Wayback Machine, TinEye |
| Watch | Follow up as sources change | AyeWatch |
Two developer use cases
Your own attack surface. theHarvester collects hostnames, email addresses, IPs and URLs for a domain from search engines, certificate transparency logs, DNS datasets and code repositories. SpiderFoot runs more than 200 modules against a domain, IP address or subnet and exports results as CSV or JSON. Both are open source and run on your machine. theHarvester's README says to point it only at targets you own or have permission to test, and that rule fits every scanner on this page.
Research on a company or a technology. Public job postings, changelogs, filings and patents often hint at a company's direction before any launch post. The SEC's EDGAR full-text search is free and spans more than 20 years of filings, so a product name or a market can be traced through a company's own disclosures.
A scan is a snapshot
Running theHarvester tells you what was true this afternoon. Next month the changelog has new entries, the careers page lists a new team, and a subreddit starts discussing an outage. AyeWatch fills that gap: describe a topic, paste a URL, or pick a public source such as an X account or a subreddit, and it sends a summary with source links when something relevant shows up.
It also fits how developers work. Paid plans include a REST API and signed webhooks, so alerts can feed your own scripts, and the AyeWatch MCP server lets an AI assistant create, update or pause monitors for you. Paid plans start at $9 a month, after a free preview.
The legal and ethical line
Public does not mean anything goes. Follow each site's terms and your local law, research organizations and technologies rather than private people, and keep scanning tools pointed at systems you own or are authorized to test.
Where to start
Pick one question, run the free lookups, and save what you find with dates. Then set up monitors for the parts that will change. For a one-time audit of your own domain, a local scan with SpiderFoot or theHarvester may be enough on its own.