ad-to-site-journey-handoff
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data such as campaign briefs, ad copy, offer facts, and market data (specified in SKILL.md). These ingestion points represent a surface for indirect prompt injection if the source content contains malicious instructions. However, the skill's capabilities are restricted to read-only analysis and the production of structured documentation like RACI tables and brief templates. It does not employ specific boundary markers or sanitization logic for these inputs, but the risk is minimized by the lack of active tools or execution environments.
- [EXTERNAL_DOWNLOADS]: The skill contains a reference to an external repository at
https://github.com/peterbamuhigire/website-skillsinreferences/landing-page-brief-template.md. This repository belongs to the skill's author (peterbamuhigire) and is intended for reporting web performance metrics. As a vendor-owned resource used for its stated purpose, it does not pose a security risk in this context.
Audit Metadata