pp-hotelist

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path is executing hotelist-pp-cli commands (e.g., rank-country, chain-compare, corridor, watch, hotel, show) which “scrape from Hotelist” and return AI-rated hotel/review/photo-verified content; that scraped outsider-authored free text is ingested into the agent via the CLI’s --json/--select output.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill requires installing and running remote code via the provided installer commands (e.g., "npx -y @mvanhorn/printing-press-library install hotelist --cli-only" and "go install github.com/mvanhorn/printing-press-library/library/travel/hotelist/cmd/hotelist-pp-cli@latest"), which fetch and execute external code that the skill depends on at runtime.