// the find
snooppr/snoop
Snoop — инструмент разведки на основе открытых данных (OSINT world)
Snoop is a Python username-enumeration tool. You give it one or more handles and it checks them against a list of roughly 5,400 sites, with optional geo/IP and Yandex plugins on top. It is aimed at OSINT practitioners, pentesters and CTF players, and it ships prebuilt Windows and Linux binaries so it can run without a Python install.
The site database is a plain list in websites.md, so you can see exactly what gets checked, diff it between releases, and fork it. Most of the tool's value sits in those entries, and that is more auditable than most closed OSINT tools. Region filters (-i/-e with country codes) and per-site selection (-s) let you narrow a run to the sites that matter, which keeps a scan from spending minutes on timeouts. Concurrency is tunable (--pool up to 300, -t per request) and a batch mode (-u) takes a file of handles, so it scales from one name to thousands without a wrapper script. The troubleshooting section is unusually candid. It lists flaky sites, regional false positives and what censorship looks like from the client side, which is operational knowledge most tools never write down.
The headline comparison (about 5,400 sites against Sherlock's 400 and WhatsMyName's 600) counts database entries, not verified hit rates. The README itself says sites go down, change their responses and return false positives, so the count tells you little about accuracy. Measure it against a known set of accounts before you trust a run. The complete database sits behind a 'full version' and donations, so the public list is a subset you cannot fully audit. That is a business decision, but it means you cannot know the exact set you are running. The README is marketing-heavy and mostly Russian. Lines like 'the most powerful OSINT software' and the NSA aside are not evidence, and the English README is a separate, thinner file. Much of the troubleshooting is tied to Russian ISP censorship, which is irrelevant to most users. Install is documented as Python 3.7+ with requirements.txt, but the primary path is prebuilt release binaries, and the README does not explain how those are verified against the source. The history was also rewritten in December 2024, so anyone pinning commits should know that.