ap2-human-not-present-flow

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

• Third-party content exposure detected (low risk: 0.10). The workflow fetches documentation and GitHub releases from specified URLs, which are first-party or selected technical sources rather than monitored outsider-authored feeds.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs fetching external specifications and release notes from ap2-protocol.org and GitHub during setup to govern implementation behavior, but lacks verifiable first-party publisher provenance for the unofficial protocol domain, making it an unverified operational dependency.