messages-ops

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to locate one-time codes in messages and return "message summary or code" in its output, which requires outputting secret OTP values verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). The “Messages Ops” runtime workflow ingests message/thread contents from a live messaging surface (e.g., local iMessage/DMs or a browser-gated message surface) that a user/outside party can author, because it explicitly requires resolving the thread then reading the latest inbound to identify open loops and retrieve one-time codes.