product-marketing

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill follows best practices for context management within an AI agent's environment.
  • [COMMAND_EXECUTION]: The skill instructs the agent to read the codebase (README, package.json, landing pages) to auto-draft marketing materials. This is a standard and safe operation for a development-oriented AI agent.
  • [DATA_EXPOSURE]: The skill creates and reads a local file at .agents/product-marketing.md. It does not attempt to access sensitive system files like SSH keys or environment secrets.
  • [EXTERNAL_DOWNLOADS]: No external network requests or script downloads were identified. All operations are performed on the local filesystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 07:25 AM
Security Audit — agent-trust-hub — product-marketing