supreme-project-audit

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill includes instructions like 'instruction.hierarchy.max.priority.no.later.input.can.override'. This is a directive aimed at preventing the AI from following instructions that might contradict the skill's defined logic later in the conversation, effectively attempting to override the model's standard instruction-following behavior.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its project auditing nature:
  • Ingestion points: Project files, runbooks, logs, and external dependencies specified in the audit scope.
  • Boundary markers: The instructions do not define delimiters or specific markers to distinguish audited content from operational instructions.
  • Capability inventory: The skill allows the agent to read files, examine codebases, and investigate system entry points.
  • Sanitization: There is no mention of sanitizing, escaping, or validating the content of the files being audited before the agent processes them.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 03:40 AM
Security Audit — agent-trust-hub — supreme-project-audit