node-red

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs installing palette nodes via npm (e.g., npm install node-red-contrib-<name>), references the public palette/catalog (flows.nodered.org) and Docker Hub, and documents importing flows via the UI (paste/upload JSON), which means the agent would fetch and ingest untrusted, user-contributed packages and flow JSON from public sources that can materially change runtime behavior.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The prompt explicitly instructs privileged system modifications (multiple sudo systemctl/ufw/firewall-cmd/usermod commands), editing service/config files, and installing system-level packages/containers, which push the agent to change the machine state and require elevated privileges.