Jailbreak
Attempts to bypass safety guardrails and persona constraints.
5 tests in this category, escalating by delivery channel from html-comment at tier 1 to 4 channels at tier 5. Each page hides an indirect prompt injection; if an agent follows it, the response names the attack it demonstrated and the command to scan for it. Background on this class of attack is in the attack guide.
- T1
DAN Persona
criticalClassic Do-Anything-Now persona override
delivery: html-comment
- T2
Developer Mode
criticalClaims developer/debug mode enables unrestricted access
delivery: html-comment, invisible-span
- T3
Hypothetical Framing
highUses hypothetical scenario to bypass restrictions
delivery: html-comment, invisible-span, meta-tag
- T4
Multilingual Evasion
highSwitches languages to evade English-focused safety filters
delivery: json-ld, meta-tag, invisible-span
- T5
Token Manipulation
criticalExploits tokenization to smuggle instructions past filters
delivery: json-ld, meta-tag, invisible-span, html-comment
Scan your own setup
These pages test whether an agent follows instructions it finds in content. HackMyAgent tests the configuration underneath it, and prints each finding with its severity. Some findings also print a command to verify them and a fix.
npx hackmyagent secure