The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses npx to execute administrative commands from the n8nac and @n8n-as-code/n8n-manager packages. These operations include managing workflow synchronization, inspecting instances, and generating technical documentation.\n- [EXTERNAL_DOWNLOADS]: Usage of npx --yes ensures that necessary n8n management utilities are automatically fetched from the npm registry if they are not already available in the local environment.\n- [CREDENTIALS_UNSAFE]: The skill explicitly defines safe handling of sensitive information. It instructs the agent to use file-based credential creation instead of passing secrets in shell arguments and strictly forbids printing API keys or secret values in the user interface.\n- [PROMPT_INJECTION]: The skill includes instructions for autonomous operation, such as performing non-interactive setup tasks without asking the user. While identified as potential concealment by static detectors, the context demonstrates this is intended for operational efficiency, as the agent is simultaneously instructed to clearly report all concrete actions and results to the user.\n- [SAFE]: The skill mitigates risks of indirect prompt injection from external data (e.g., workflow definitions or execution logs) by mandating the use of schema validation commands (node-info, node-schema) as the absolute source of truth.

n8n-architect