compensation-benchmarking

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md Data Sources explicitly instructs using "web research, public salary data, and user-provided context" and to "pull verified benchmarks," meaning the agent will fetch and interpret open/public third-party content (e.g., salary sites) that can materially influence compensation decisions and thus could enable indirect prompt injection.