Maigret
github.com/soxoj/maigret- Category
- Cybersecurity
- Rank
- No. 306Tools index
- Type
- APP
- Builder
- soxoj
- GitHub
- 37.6k stars
- Latest release
- nightly-main
- Date
About
Maigret is an OSINT tool that builds a dossier on a person from a username alone, checking thousands of sites for accounts and gathering the available information from their pages — no API keys required.
What it does
Maigret runs concurrent profile checks across a configurable site database, classifies each result, extracts identity clues, and can follow newly discovered usernames. Results can be filtered by country or category and exported as documents, structured data, or interactive graphs.
Why it's ranked here
Its breadth, recursive discovery, and unusually broad export surface make it practical for real investigations. The site database updates automatically but retains an offline fallback. Block detection and explicit error states also help distinguish failed checks from absent accounts, though protected sites remain an unavoidable source of uncertainty.
What's good
The default scan favors higher-traffic sites, while tags and full scans provide tighter control. Maigret supports terminal use, Python embedding, and a browser interface. HTML, CSV, text, JSON, graph, XMind, PDF, and Neo4j outputs let the same findings feed human review, automation, and graph analysis.
Tradeoffs
CAPTCHAs, censorship, DNS failures, timeouts, and bot protection can prevent reliable checks; Maigret detects several such conditions but only partially bypasses them. PDF output needs an optional package set plus system graphics libraries on Linux and macOS. AI summaries require a separate OpenAI-compatible API key and send the generated report to that service.
How to use it well
Use Maigret for OSINT triage when a username is the strongest starting clue. Begin with the default high-traffic set, narrow by tags when context is known, then inspect extracted links and graph relationships before expanding recursively. Treat matches as investigation leads, not standalone proof that separate accounts belong to one person.
Technical notes+
pyproject.toml defines a Python 3.10+ Poetry package, the maigret CLI entry point, an MIT license, optional PDF dependencies, and development tooling for pytest, coverage, flake8, mypy, and Black. maigret/sites.py models engines, sites, URL matching, ranking, mirrors, and identifier types. maigret/result.py separates claimed, available, unknown, and illegal outcomes. maigret/errors.py recognizes protection and network failures and applies configurable significance thresholds. maigret/report.py generates multiple report and graph forms, while guarding PDF image resolution against local, private, multicast, and reserved targets. maigret/ai.py streams summaries from an OpenAI-compatible chat-completions endpoint.
Observed
- License
- MIT License
- Primary language
- Python 3.10 or newer
- Installation
- PyPI package, source install, Docker images, and a standalone Windows executable
- Interfaces
- Command-line interface, embeddable Python library, and built-in web interface
- Platform support
- Package metadata classifies the project as operating-system independent
- Testing and quality
- The Makefile defines pytest coverage, flake8, mypy, and Black targets
Read from README.md, Makefile, pyproject.toml, wizard.py, maigret/ai.py, maigret/sites.py, maigret/utils.py, maigret/errors.py, maigret/notify.py, maigret/report.py, maigret/result.py, maigret/submit.py.
Tags
Tech Stack
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.