analytics-architecture

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Prompt Injection (SAFE): No attempts to bypass agent constraints or ignore instructions were detected. The skill maintains a consistent professional persona.
  • Data Exposure (SAFE): The skill explicitly includes logic to detect and prevent the tracking of sensitive data (email, SSN, phone numbers). No hardcoded secrets were found in the provided files.
  • External Downloads & RCE (SAFE): The skill does not download external scripts, install packages, or execute arbitrary code.
  • Indirect Prompt Injection (LOW): The skill is intended to process untrusted user code for 'Diagnosis' and 'Review' tasks. While this is a data ingestion surface, the provided instructions prioritize privacy and security compliance, reducing the risk of accidental exploitation.
  • Metadata Poisoning (SAFE): All metadata is relevant to the domain of product analytics. A reference to a missing file 'sharp_edges.md' was noted in the instructions, which is a minor configuration issue rather than a security threat.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:15 PM