competitor-analysis

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md instructs the agent to ingest “supplied URLs, screenshots, customer notes, sales calls, reviews, pricing pages… app listings, or marketplace pages,” which at runtime typically requires fetching/parsing outsider-authored web content or user-provided external notes into the LLM context (indirect prompt-injection risk).