⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⣤⣤⣤⣄⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣤⣶⣿⣿⣿⣿⣿⣿⣿⣷⣄⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣴⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣆⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⣰⡆⠀⠀⠀⠀⠀⠀⠀⢀⣴⣿⣿⣿⣿⡿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⡟⢹⡀⠀⠀⠀⠀⠀⣠⣾⣿⣿⡿⠋⠁⠶⠿⠿⠿⢿⣿⣿⣿⣿⣿⣿⣿⣄⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⣧⣴⣦⣄⡀⠈⣿⣿⣿⡟⠀⠀⠀⠀⠀⠀⠀⠀⠈⠻⣿⣿⣿⣿⣿⣿⠷⣆⣀⣀⣀⠀⠀⠀⠀
⠀⠀⣞⣠⣶⣿⣿⣿⣿⣿⣿⣿⣿⡿⠀⠀⠀⢀⣴⠟⢻⡄⠀⠀⠀⠈⢿⢻⣿⣿⣿⠀⠀⠉⠉⠉⠳⡄⠀⠀ _ _
⠀⠀⣿⣿⠿⣿⣿⡟⢸⣿⣿⣿⣿⠃⠀⠀⣠⠞⠁⠀⠀⠛⢦⡀⠀⠀⠀⠀⣿⣿⣿⠀⠀⠠⡀⠀⠀⠙⡄⠀ | | | |
⠀⢰⣿⡋⠀⠙⠋⠁⢺⣿⣿⣿⣿⡄⠀⡴⠁⠀⠀⠀⠀⠀⠀⠈⢷⡀⠀⠀⣿⣿⣿⠀⠀⠀⢿⠀⠀⠀⠹⠀ _ __ __ _| | _____ __| |
⠀⠸⠉⠀⣀⣀⣀⡀⢸⣿⣿⣿⣿⢃⠌⠀⠀⠀⠀⠀⠀⠀⠀⠀⠈⡇⠀⠀⣿⣿⡏⡀⠀⠀⢸⡁⣤⠀⠀⡀ | '_ \ / _` | |/ / _ \/ _` |
⠀⠃⠀⡼⠁⠀⠀⠁⠈⣿⣿⣿⣯⡎⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⡇⠀⢸⣿⣿⢰⣿⡆⠀⢸⣿⣿⠀⠀⡇ | | | | (_| | < __/ (_| |
⣀⡿⠙⠁⠀⠀⠀⠀⠀⠘⣿⣿⣿⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠰⢧⡀⠘⢿⣿⡆⠋⠀⢀⣾⣿⣿⡷⢀⡇ |_| |_|\__,_|_|\_\___|\__,_|
⠁⠀⠀⠀⠀⠀⣤⣀⠀⣇⠸⣿⣿⣧⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⡙⢦⣼⣿⣿⡄⢀⣼⣿⣿⣿⣧⣾⠃
⠀⠀⠀⠀⠀⠀⠹⣿⣷⣾⣆⢹⣿⣿⣠⣦⠀⠀⠀⠀⠀⠀⢀⠀⢄⣾⣿⣦⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⠏⠀
⠀⠀⠀⠀⠀⠀⠀⢹⣏⠻⠟⠀⢻⣿⣿⣿⠀⠀⠀⠀⢀⢀⣀⣤⣤⣭⣙⣿⣿⣿⣿⣿⣿⣿⣿⣿⡿⠃⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⢻⡀⠀⠀⠀⢿⠋⣿⡇⠀⠀⢀⡾⠟⠛⣿⡿⠙⢻⣿⣿⠿⠿⠿⠿⠿⠋⠁⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠃⠀⠀⠀⠀⠀⠘⣿⠀⠀⠈⠀⠀⣰⠏⠀⢀⣿⣿⠏⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢹⡄⠀⠀⠀⠠⠋⠀⠀⣼⣿⣷⠞⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢧⠀⠀⠀⠀⠀⠀⠀⠉⣼⠋⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠘⠀⠀⠀⠀⠀⠀⠀⠠⠃⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
Username correlation and identity research framework.
NAKED is a personal project that does, as the name suggests, what it says: it strips an identity down to what's actually verifiable. The core is built to automate that search — collecting and correlating an identity's traces across platforms — so you don't have to do it by hand.
Given a username, NAKED queries multiple providers in parallel and returns not just whether an account exists, but how much you should trust that result.
Every result carries a status: FOUND, NOT_FOUND, or ERROR. "Doesn't exist" and "couldn't verify" are not the same thing, and treating them as equivalent produces false negatives. When a result is FOUND, NAKED also computes a NAKED Score with the reasons behind the number — not a value pulled out of nowhere.
github
status : FOUND
naked score : 100/100
reasons : Official API, Username exact match, Public profile, Profile URL verified
reddit
status : ERROR
error : Reddit returned 403 (suspended, private, or blocked)
naked/ core/ engine, plugin loader, cache, logging intelligence/ scoring engine score.py ScoreCalculator rules.py per-provider scoring rules models/ pydantic models (SearchResult, profiles, score) providers/ one folder per source github/ reddit/ examples/dummy/ main.py CLI entrypoint (typer)
Providers are auto-discovered: drop a folder under providers/ with a provider.py defining a Provider subclass, and the PluginManager loads it at runtime. No manual registration needed.
Requires Python 3.14+ and uv.
git clone https://github.com/SnakeFirts/NAKED.git
cd NAKED
uv syncuv run python main.py <username>
uv run python main.py --help| Provider | Source | Auth required |
|---|---|---|
| github | api.github.com (official API) | No |
| reddit.com/user//about.json | No |
More providers are added incrementally.
Each provider defines its own rules in naked/intelligence/rules.py. Points are awarded for things like using an official API, an exact username match, a public profile, or a verified account — the total is capped at 100. The calculator lives independently from the providers, so scoring logic can evolve without touching provider code.
uv run pytest tests/ -vTests cover scoring logic and error classification (including mocked rate-limit and ambiguous-block scenarios) without depending on live network calls.
[x] Plugin-based provider architecture [x] GitHub provider [x] Reddit provider [x] Intelligence scoring engine [x] FOUND / NOT_FOUND / ERROR result classification [x] CLI with username argument [ ] Additional providers (scraping-based) [ ] Cross-provider correlation / overall identity score [ ] Rich-formatted terminal output
TBD