ad-creative-intelligence

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and analyzes untrusted data from external ad platforms.
  • Ingestion points: Competitor ad copy, headlines, and descriptions are scraped from Meta and Google Ad Libraries (SKILL.md).
  • Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" warnings for the agent when processing the scraped content.
  • Capability inventory: The skill executes shell commands via python3 and performs file-write operations to the local file system (SKILL.md).
  • Sanitization: No sanitization or escaping of the scraped ad content is performed before it is analyzed by the agent.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute local Python scripts using shell commands.
  • Evidence: The instructions include shell commands like python3 skills/meta-ad-scraper/scripts/scrape_meta_ads.py --domain <competitor_domain> which interpolate user-supplied domain names directly into the command line arguments (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 05:17 PM