pp-visit-detroit-blog

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime path is executing visit-detroit-blog-pp-cli to read Detroit blog article bodies from its local SQLite store populated by sync from Algolia/upstream content; that ingested article text is outsider-authored free-form prose (not user-authored) and can be returned to the agent via --agent/--json outputs.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill's required install steps invoke remote installers that fetch and execute code at runtime (e.g., "npx -y @mvanhorn/printing-press-library install visit-detroit-blog --cli-only" and "go install github.com/mvanhorn/printing-press-library/library/travel/visit-detroit-blog/cmd/visit-detroit-blog-pp-cli@latest"), so it depends on external content that is downloaded and executed during setup.