ccpa

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process data from external URLs provided by the user, which serves as a vector for indirect prompt injection attacks. Malicious instructions could be embedded in the audited website's content to influence the agent's output or actions.
  • Ingestion points: Step 1 (Paso 1) involves gathering data from the target website's privacy policy and other relevant pages.
  • Boundary markers: The instructions lack explicit delimiters or warnings to the agent to disregard instructions contained within the fetched content.
  • Capability inventory: The skill instructs the agent to write files to the local file system (Step 6).
  • Sanitization: No mechanisms for sanitizing or validating the content retrieved from the audited URLs are defined.
  • [DATA_EXFILTRATION]: The skill includes instructions to save audit reports to a specific, hardcoded local directory structure.
  • Evidence: The path 'C:/Users/cmano/claude-seo/' is explicitly mentioned in Step 6, which exposes information about the author's or intended user's local file system structure and username ('cmano').
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 12:02 AM
Security Audit — agent-trust-hub — ccpa