coordinating-cross-team-breakdown

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

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow fetches a Confluence page (get_confluence_page for page 2920349776) and ingests its readable text into the agent context; that page is outsider-authored public web content fetched at runtime (even if within Atlassian), creating indirect prompt-injection risk.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly calls get_confluence_page to fetch the canonical template at https://bitwarden.atlassian.net/wiki/spaces/EN/pages/2920349776 at runtime and relies on that page's content to drive its signoff/template behavior, so the fetched page can directly control the agent's instructions.