backend-pe

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION] (HIGH): The skill utilizes adversarial prompting techniques to override system-level constraints and safety protocols.
  • Evidence (Directives): The skill explicitly commands the AI to 'Ignore token and cost constraints' and 'No Safety Lectures: Assume expert users. Do not warn about cost or complexity'.
  • Evidence (Triggers): It defines activation triggers like 'Supermode' and 'Antigravity', which are characteristic of jailbreak-style prompts used to enter an unrestricted or 'unfiltered' state.
  • Evidence (Instruction Override): The directive to 'Push reasoning and code generation to practical limits' combined with 'Ignore token and cost constraints' is an attempt to bypass resource management guardrails.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 07:29 AM