ollama

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

• Third-party content exposure detected (high risk: 0.70). The ollama service exposes an HTTP API (e.g., /api/tags and chat/generate endpoints) and at runtime it will ingest user-supplied request bodies (free-form prompts) into the model context; those prompts are outsider-authored when they come from other users/clients than the operating user.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill describes system-level installation (extracting a binary to /usr) and installing/managing a supervisord service and containers (modifying system service files and state), which are privileged operations that change the machine state and may require sudo, so it should be flagged.