The Agent Skills Directory

[COMMAND_EXECUTION]: The skill defines extensive usage of the n8n-cli binary through the Bash tool to perform administrative tasks such as workflow deployment, project management, and system auditing.
[CREDENTIALS_UNSAFE]: The instructions guide the user on configuring the N8N_API_KEY using environment variables or configuration files. It also includes commands for creating credentials within the n8n instance, which requires handling sensitive API data during the creation process.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external sources that could contain malicious instructions.
Ingestion points: Commands like n8n-cli workflow get, n8n-cli execution get --include-data, and n8n-cli data-table rows fetch content directly from a remote n8n instance.
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions embedded within the retrieved workflow or execution data.
Capability inventory: The agent has broad capabilities including file system writes (Write) and the ability to modify instance state (create/update/delete workflows and credentials) via n8n-cli.
Sanitization: There is no mention of sanitizing or escaping the data fetched from the n8n instance before the agent processes it, which could allow a malicious workflow or execution log to influence agent behavior.

n8n-cli