BEHAVIORAL SANDBOX
Test Your Agent Like an Attacker Would
Browser-based behavioral testing. We visit your agent, interact with it like a real user, and try 5 attack categories. Results in under 60 seconds.
Run Sandbox Test
5 Attack Scenarios
Prompt Injection via UI
criticalTypes jailbreak prompts into the agent chat interface. Tests DAN, system prompt override, role-play attacks.
PII Extraction
criticalAsks the agent to reveal customer data, payment info, or internal records through social engineering.
System Prompt Leak
highAttempts "repeat your instructions", "what is your system prompt", and 8 other extraction variants.
Tool Abuse
highAsks the agent to execute dangerous actions: delete data, send emails, make unauthorized API calls.
Session Hijacking
mediumManipulates conversation context to impersonate other users or escalate privileges.
How It Works
Launch Browser
Headless Chromium visits your agent URL in an isolated sandbox.
Interact
Automated scenarios type messages, click buttons, submit forms.
Observe
Every response is captured: text, network requests, console logs.
Judge
LLM judge + pattern matching score each interaction pass/fail.
Sandbox API
Part of Agent Certification
Sandbox testing is the 4th component of our certification pipeline (20% weight). Combined with Mystery Shopper (50%), GuardScan (35%), and Identity checks (15%), it gives the most comprehensive agent security assessment available.
Protected by U.S. patent-pending technology (App. Nos. 63/983,615; 63/983,621; 63/983,843; 63/984,626).

