“"Seems incredibly prone to false positives." — AdmiralAsshat, Hacker News”
You know that feeling when you search the same username across site after site, copy links into notes, and still wonder what you missed. You also know the next problem: a found profile often points to another handle, so the search keeps branching while your browser tabs pile up. Maigret tackles that by turning one username into a broad scan, a second pass on linked identities, and a report set you can review.
You give Maigret a username, and it loads its site database and checks the top 500 ranked sites unless you widen the scope. When it finds a profile, it parses page data such as linked accounts and other identifiers, then it can launch more searches from those new leads. Think of it like starting with one name on a sticky note, then following every breadcrumb that points to another account. At the end, you get terminal output plus report files such as HTML, PDF, JSON, CSV, TXT, XMind, or a graph view.
If you run investigations, trust-and-safety reviews, fraud checks, or online identity research, you get the most value here because you can treat Maigret as a lead-finding layer and review the hits yourself. If you already live in Python, the async API also lets you fold it into your own workflow. It is not for you if you need clean, court-ready identity matching from a single run or if you work from a blocked VPS and cannot change networks.
You should explore it if you need breadth, report generation, and a Python-friendly OSINT component. It looks production-proven as a maintained building block because the repo stays active, ships releases, and the docs say commercial tools build on it. You should not treat it as automatic truth because the docs, issues, and biggest HN thread all point to false positives, WAF trouble, and some docs-to-package drift.
Deep-dive insight, Easy and Pro modes, plus action playbooks — the full breakdown is one tap away.