Persona

Persona is a custom wordlist generator — it turns a target's personal details into a list of likely passwords and usernames. It is the same category of tool as CUPP and Username Anarchy, built for authorized penetration testing, auditing your own accounts, and password-awareness education.

It ships with an Uzbek module that understands local spelling habits (oʻ/gʻ, Latin↔Cyrillic) and common local password patterns.

Supported platforms: Windows and Linux (Ubuntu / Debian), on amd64 and arm64. It is a single static binary with no runtime dependencies.

⚠️ Ethical Use Persona never attacks, connects to, or cracks anything. It only reads the data you give it and writes .txt files. Use it only against systems and accounts you own or are explicitly authorized to test. You are responsible for complying with all applicable laws. The authors accept no liability for misuse.

---

Features

• Interactive TUI (Charm: Bubble Tea / Lip Gloss / Bubbles) with an ASCII banner, guided prompts, and a progress spinner.
• Pipeline: Profile → tokens → mutations → combinations → dedupe → filter → rank.
• Mutations: case (ali/Ali/ALI), leetspeak (a→@/4, o→0, i→1/!, …), reverse, truncation.
• Combinations: token + separator + token, token + year, token + suffix.
• Uzbek normalization + Latin↔Cyrillic transliteration.
• Username templates (first.last, flast, f.last, first+year, …).
• Policy filters: --min, --max, --require-digit, --require-symbol.
• Explosion guard: --max-words, --depth.
• Config-file and flag-driven non-interactive runs for CI.

Install

Replace Yescript with your own GitHub username/org if you fork this.

1. Linux (Ubuntu / Debian) — one-liner

No apt packages required; the script downloads a prebuilt binary and adds it to your PATH.

curl -fsSL https://raw.githubusercontent.com/Yescript/persona/main/install.sh | bash

2. Windows — one-liner (PowerShell)

irm https://raw.githubusercontent.com/Yescript/persona/main/install.ps1 | iex

3. Manual binary

Download the asset for your OS/arch from the Releases page.

Linux:

tar -xzf persona_linux_amd64.tar.gz
chmod +x persona
mv persona ~/.local/bin/

Windows: unzip persona_windows_amd64.zip and move persona.exe into a folder on your PATH.

4. Go install (any OS with Go 1.22+)

go install github.com/Yescript/persona/cmd/persona@latest

5. From source

git clone https://github.com/Yescript/persona.git
cd persona
make install          # Linux/macOS
go build -o persona.exe ./cmd/persona   # Windows

Usage

Interactive (default):

persona

With flags:

persona --name Ali --surname Valiyev --birth 1999 -o wordlist.txt

From a config file:

persona --config profile.yaml

Usernames only:

persona --usernames-only

Flags

Flag	Default	Description
--name		First name
--surname		Surname
--birth		Birth year (e.g. 1999)
--config		Load a profile from a YAML file (non-interactive)
-o, --output	wordlist.txt	Output wordlist path
--usernames	false	Also write usernames.txt alongside the wordlist
--usernames-only	false	Only generate usernames
--usernames-output	usernames.txt	Usernames output path
--min	0	Minimum length (0 = no limit)
--max	0	Maximum length (0 = no limit)
--require-digit	false	Keep only candidates with a digit
--require-symbol	false	Keep only candidates with a symbol
--max-words	100000	Cap total generated words
--depth	2	How many tokens may combine (1 or 2)
--leet	true	Enable leetspeak mutations
--symbols	true	Allow symbol suffixes
--years	true	Append year tokens
--cyrillic	false	Also emit Cyrillic transliterations (Latin-only by default)
--reverse	true	Include reversed tokens (--reverse=false to disable)
--truncate	true	Include truncated tokens (--truncate=false to disable)
--stdout	false	Also print results to stdout
--version		Print version and exit

The Uzbek module

Many Uzbek users build passwords from Latinized names plus numbers, phone tails, favorite clubs, or religious words. Persona's uzbek package:

• Splits free-text fields into clean tokens (so black, white, ko'k becomes three tokens, never one comma-joined string).
• Normalizes oʻ/o' → o, gʻ/g' → g, and strips stray apostrophes.
• With --cyrillic, produces Latin→Cyrillic variants (e.g. Akmal → Акмал), keeping digraphs like sh, ch, ng. Transliterations that would be broken (e.g. English white) are dropped, not emitted. Output is Latin-only by default, since most Uzbek users type Latin passwords.
• Seeds common local passwords and keyboard patterns.

Output

• wordlist.txt — candidate passwords (primary output).
• usernames.txt — candidate usernames (when enabled).
• Final stats: token count, generated count, file size.

Development

make build   # build ./persona
make test    # run unit tests
make fmt     # gofmt
make vet     # go vet

This repo targets Go 1.22+. Run go mod tidy once to fetch dependencies and generate go.sum.

Contributing

Issues and PRs welcome — especially additional Uzbek tokens, transliteration fixes, and username templates. Keep packages small and gofmt/go vet clean, and add tests for new generation logic.

License

MIT — see LICENSE.