The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown files and metadata configuration. It contains no executable scripts, system commands, or network-enabled operations.
[NO_CODE]: No code or scripts were found in the skill files. All logic is provided through natural language instructions for the AI agent.
[INDIRECT_PROMPT_INJECTION]: The skill processes external visual references and project contracts to generate art direction rules, which represents an attack surface for instructions embedded in data.
Ingestion points: Reference files, existing art contracts, and owner decision documents described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for external data.
Capability inventory: The skill instructs the agent to save creative artifacts to the {agentRoot}/creative/ directory and use other specialized design skills.
Sanitization: No explicit sanitization or input filtering is mentioned, as the workflow focuses on artistic interpretation.

art-direction