NYC

Quality Assurance

Warn

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: MEDIUMNO_CODEPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (MEDIUM): The skill identifies as a code-quality enforcer and has the capability to modify files. This creates a significant risk where malicious code processed by the skill could inject instructions to the agent. Evidence: Ingestion points: Project files; Boundary markers: Absent; Capability: modifies_files; Sanitization: None.
  • [No Executable Code] (INFO): Only the manifest was provided for analysis. The lack of logic scripts or prompts limits the assessment to defined capabilities.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 10:40 PM