shopping-campaign-audit

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests data from Google Ads reports and web tools, which introduces a potential surface for indirect prompt injection. However, this is inherent to its primary function as an audit tool.
  • Ingestion points: External data is ingested through GAQL query results (such as product_title) and the output of WebFetch and WebSearch tools.
  • Boundary markers: The instructions do not define specific delimiters to separate untrusted data from instructions, though the task is structured and analytical in nature.
  • Capability inventory: The skill uses tools to read Google Ads data and generate findings but does not possess destructive capabilities (e.g., file-system writes or unauthorized command execution).
  • Sanitization: Standard sanitization for product titles or web content is not explicitly defined in the prompts.
  • [SAFE]: All external tools (mcp__cogny__*) and referenced URLs (cogny.com) are owned and maintained by the vendor (Cogny AI), representing legitimate vendor functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 11:31 AM
Security Audit — agent-trust-hub — shopping-campaign-audit