Find and prevent secrets and prompt injection attacks in AI coding agent sessions
Stage
Not given
Not given
Founded
Not given
Not given
Based in
CA
Not given
Pricing model
Freemium
Open source
Pricing
Starter (Free): 5 scam checks every month, read any QR code as often as you like, private scan history. Pro: $7/month for 30 scam checks every month, unlimited QR code generation, changeable QR codes, and scan analytics.
Not given
Key features
Instant QR code scanning under a second
Live URL threat analysis before opening links
Detects fake websites and copycat payment pages
Works with blurry, faded, and stickered QR codes
No tracking or passwords required
Private scan history for registered users
99.9% uptime commitment
Plain language security warnings
Scan recorded AI agent sessions for known credentials
Guard against prompt injection attacks
Local console with audit trails at 127.0.0.1
Support for multiple AI coding agents including Claude Code, Cursor, and Codex CLI
Default policy with 21 rules across 19 action classes
Test policy against replayed attack scenarios
No model or network in the guard's hot path
What makes it different
Checks links before you tap, not after
Designed specifically for family protection
Reads damaged QR codes that other scanners miss
Deterministic policy-based enforcement rather than model-based decisions
Runs locally with no telemetry and no account required
Open source and free forever under Apache-2.0
Replayed 36 real attack scenarios with zero bypasses
QR Shield
Stroq