tryclear 3.0.0

TryClear — Semantic image threat evaluation from the command line

TryClear CLI

A picture is worth a thousand words. Attackers know that. Your security tools don't.

TryClear evaluates images for semantic threats — phishing, brand impersonation, QR code attacks, hidden page techniques, and visual prompt injection.

Setup

# 1. Install
pip install tryclear

# 2. Authenticate (grab your API key from tryclear.io/settings)
tryclear auth

# 3. Evaluate
tryclear eval suspicious-screenshot.png

That's it. Forensic verdict in your terminal.

Commands

Command	Description
tryclear auth	Store your API key
tryclear eval <file>	Evaluate an image, get a forensic verdict
tryclear eval --json <file>	Structured JSON output for scripting
tryclear status <id>	Check evaluation status
tryclear report <id>	Fetch full forensic report

Supported Formats

PNG, JPG, JPEG, BMP, TIFF, SVG, WebP, HEIC/HEIF

Example

$ tryclear eval bank-document.tiff

  ◆ TryClear v2.0.0

  bank-document.tiff · 269 KB · image/tiff

  ╭──────────────────────────────────────────────────╮
  │   MALICIOUS · High Risk                          │
  ╰──────────────────────────────────────────────────╯

  This image impersonates Bank of America using a
  hidden page technique to evade security scanners.

  What we found
    Page 1 appeared clean, but hidden content on
    Page 2 revealed the threat.

    • Bank of America branding (0.98)
    • QR code → hxxps://mail[.]google[.]com/mail/u/0/

  MITRE T1566

  Full report → https://tryclear.io/report/ana_demo

CI/CD Integration

tryclear eval ./assets/ --fail-on malicious --exit-code 1

Exit Code	Meaning
0	CLEAR
1	SUSPICIOUS
2	MALICIOUS
3	Error

Auth

The CLI resolves your API key in this order:

1. TRYCLEAR_API_KEY environment variable
2. ~/.config/tryclear/config.json
3. ~/.config/clear/config.json

Learn More

tryclear.io