aspire-mailpit-integration

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime path ingests outsider-authored free text via Mailpit’s API responses (e.g., MailpitClient.GetMessagesAsync() calls /api/v1/messages and reads HtmlBody/Subject into LLM context if the agent uses those fields), where the email content originates from the application/users rather than the operating user.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill calls builder.AddContainer("mailpit", "axllent/mailpit"), which at runtime pulls and runs the remote container image "axllent/mailpit" (a required external binary) and thus executes remote code from that external source.