survey-generator

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

• Third-party content exposure detected (high risk: 0.90). This skill's required workflow (SKILL.md Step 1 "Read the anchor") explicitly instructs the agent to WebFetch a public source_url (e.g., arXiv surveys, GitHub awesome-lists, blog posts) and extract subtopics/cited papers that are then used to build the research_bundle and drive the LLM generator, so untrusted third‑party content directly feeds and can influence subsequent actions.