bonus-stack-builder

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

• Third-party content exposure detected (high risk: 1.00). The skill's Workflow Step 1 explicitly instructs using web_search and researching queries like "[product] complaints" or "[product] struggles" (SKILL.md → "Step 1: Gather Context"), which requires fetching and interpreting open/public user-generated content (reviews, forum posts, complaints) that the agent will read and use to decide and design bonuses.