market-ads
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external web content from user-provided URLs to generate advertising campaigns. This behavior presents an attack surface for indirect prompt injection.\n
- Ingestion points: External content is fetched and analyzed as part of the core workflow defined in
SKILL.md.\n - Boundary markers: The instructions include safety constraints for output but do not explicitly define boundaries to ignore instructions found within external content.\n
- Capability inventory: The agent has the ability to write files (
AD-CAMPAIGNS.md) and display results in the terminal.\n - Sanitization: There is no documentation of sanitization or filtering applied to the content retrieved from the target URL before processing.
Audit Metadata