pp-drudgereport

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

• Third-party content exposure detected (medium risk: 0.65). SKILL.md describes commands like drudgereport-pp-cli page and drudgereport-pp-cli feed that fetch external/public Drudge content and an unofficial RSS mirror at runtime; if the agent then uses that fetched HTML/RSS-derived text in its LLM reasoning context, it would be outsider-authored free text via the runtime network-fetch → CLI output → agent LLM context path.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill instructs installing the CLI via a remote Go module which will fetch and execute code at install time (go install github.com/mvanhorn/printing-press-library/library/media-and-entertainment/drudgereport/cmd/drudgereport-pp-cli@latest), so this is a runtime external dependency that can execute remote code.