summarize

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly fetches and ingests content from public URLs and YouTube (see SKILL.md "Summarize" and the quick-start examples), meaning untrusted user-generated third-party content is read and used to produce summaries that could embed instructions influencing the agent's subsequent outputs or actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The summarize skill fetches arbitrary user-supplied URLs at runtime (e.g., https://example.com and https://youtu.be/dQw4w9WgXcQ) and injects the fetched content into the model prompt/context to produce summaries, meaning remote-hosted content can directly control agent input.