video-script

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

• Third-party content exposure detected (high risk: 0.90). The skill's Step 2 "Research (If Needed)" in SKILL.md explicitly instructs the agent to use WebFetch/WebSearch and to read reference URLs, docs, and public repos provided by the user, meaning it ingests untrusted third‑party web content and summarizes/acts on it (e.g., for script content), which can enable indirect prompt injection.