The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains instructions that prevent the agent from displaying API keys or credential secrets to the user. While flagged by automated detectors as a concealment pattern, this is a security best practice designed to prevent accidental data exposure.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes external n8n community examples and local workflow files.
Ingestion points: Reads local .workflow.ts files, AGENTS.md, and downloads community examples via n8nac skills examples download.
Boundary markers: The instructions include a warning to treat schema output as the absolute source of truth and to use examples only for learning patterns, not as authority.
Capability inventory: The skill can execute shell commands via npx, write files to the workspace, and initiate network tunnels.
Sanitization: The skill utilizes local validation tools (n8nac skills validate) to check workflow integrity before processing.
[COMMAND_EXECUTION]: The skill uses npx to run lifecycle management tools for n8n instances and workflows. Potential risks associated with infrastructure creation and network tunneling are mitigated by explicit requirements for user approval before these actions are performed.

n8n-architect