hotel-booking-timing

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md:35-37 describes extracting the API’s Server-Sent Events data: payload and relaying the returned markdown verbatim (outsider-authored free text from a public web API response) into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill issues a runtime POST/curl to https://www.besttimetobookhotels.com/api/mcp and explicitly requires relaying the returned markdown verbatim, so remote content from that URL directly controls the agent's output.