Maigret

Maigret is a command-line OSINT tool that takes a single username and checks it against thousands of websites, pulling in whatever profile data it can find along the way. It's built for OSINT researchers, penetration testers, journalists, and security teams who need to map a person's online footprint starting from just a handle, with no API keys or paid data sources required.

Key features

• Wide site coverage: checks 3,000+ sites, with a default run scanning the top 500 by traffic; the -a flag scans everything, and --tags filters by category or country.
• Data extraction: pulls names, bios, links, and other details straight from profile pages and site APIs, not just "account exists/doesn't exist" flags.
• Recursive search: follows usernames and IDs discovered during a scan to run follow-up searches automatically.
• Library usage: importable as a Python package for embedding username searches in custom scripts or pipelines.
• Auto-updated database: fetches a refreshed site list from GitHub roughly once a day, falling back to the bundled database when offline.
• Tor/I2P and proxy support: routes checks through SOCKS/HTTP proxies, Tor, or I2P, useful for .onion/.i2p targets and dodging IP-based blocks.
• Multiple report formats: exports to HTML, PDF, XMind, CSV, plain text, newline-delimited JSON, an interactive D3 graph, or a Neo4j Cypher script.
• Web interface: a built-in UI for browsing results as a graph and downloading reports, runnable locally or via a published Docker image.
• AI summary mode: --ai sends findings to an OpenAI-compatible endpoint to generate a short investigation summary (likely name, location, occupation, confidence, leads).
• Cloudflare bypass (experimental): offloads JS-challenge sites to a local FlareSolverr instance for sites that need a real browser.

Ideal use cases

Maigret fits OSINT investigations, background checks, and pentesting recon where the only starting point is a username or handle. It's a good match for security researchers verifying account exposure, journalists tracing sockpuppet accounts, and analysts building link graphs between usernames across platforms. The --permute flag (generating username variants from a name) and --parse (extracting IDs from a profile URL to feed recursive search) make it useful for expanding a thin lead into a fuller picture.

It's not a fit if you need guaranteed real-identity resolution: results are probabilistic, based on public profile pages, and can include false positives or stale data. It's also not built for bulk, high-volume scanning across many usernames at once without hitting rate limits or blocks from target sites, since it relies on scraping public pages rather than official APIs. If you need verified, API-backed identity data with SLAs, a commercial identity-verification service is a better match than a scraping-based OSINT tool.

Installation

Requires Python 3.10 or higher. The fastest path:

pip install maigret
maigret YOUR_USERNAME

From source:

git clone https://github.com/soxoj/maigret && cd maigret
pip3 install .
maigret username

Docker (CLI image):

docker pull soxoj/maigret
docker run -v /mydir:/app/reports soxoj/maigret:latest username --html

Docker (Web UI image):

docker run -p 5000:5000 soxoj/maigret:web
# open http://localhost:5000

Windows users can download maigret_standalone.exe from the Releases page and run it directly or from a terminal. Cloud Shell, Colab, Binder, and Replit options are also available for running Maigret without a local install. PDF report generation is an optional extra requiring system graphics libraries: pip install 'maigret[pdf]'.

To launch the web UI locally after installing:

maigret --web 5000

Then open http://127.0.0.1:5000 and enter a username to search.