send-message

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to fetch and read private messages from other users via the OpenAnt CLI (e.g., npx @openant-ai/cli@latest messages read <conversationId> and notifications list), which are untrusted, user-generated third-party content that the agent is expected to interpret and that could influence its subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill repeatedly instructs running "npx @openant-ai/cli@latest" at runtime, which fetches and executes remote npm package code from the registry and is required for the skill to read/send messages, so the fetched content can execute code and directly control runtime behavior.