n8n-agents-official

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly recommends "inject the storage keys into the system prompt" and have tools "accept the key as a string parameter," which requires the model to see and emit secret key strings verbatim (exfiltration risk).

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow ingests outsider-authored chat text via n8n chat/webhook triggers (e.g., chatInput from @n8n/n8n-nodes-langchain.chatTrigger or Slack Trigger) and passes it into the Agent node as the user prompt (text: '={{ $json.userMessage }}' / text: '={{ $json.text }}'), making indirect prompt injection possible.