competitor-ad-intelligence
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests and analyzes untrusted data from external websites, which creates a potential surface for indirect prompt injection attacks.
- Ingestion points: SKILL.md identifies multiple points where external content is ingested, specifically in Phase 1 (Meta Ad Library), Phase 2 (Google Ads Transparency Center), and Phase 4 (Competitor Landing Pages) using the
web_searchandfetch_webpagetools. - Boundary markers: The instructions do not define delimiters or provide specific warnings to the agent to disregard instructions that may be embedded within the scraped ad copy or landing page text.
- Capability inventory: The skill's capabilities are limited to
web_search,fetch_webpage, andcurl. There are no instructions involving subprocess execution, file-system modifications, or the use of sensitive credentials. - Sanitization: The skill lacks requirements for sanitizing or validating retrieved data before it is processed by the agent for creative pattern clustering and strategic teardown analysis.
Audit Metadata