The Agent Skills Directory

[SAFE]: The skill uses the official renku CLI for project management and Lookbook operations. All external resources and instructions refer to the vendor's domain (gorenku.com) and standard shell utilities (cd, find) for file inspection.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from user instructions and Inspiration folders to generate Lookbook JSON documents, which represents a potential injection surface.
Ingestion points: The skill reads user direction, screenplay context, and external Inspiration analysis results from the project's inspiration folders as described in SKILL.md and references/using-inspiration-sources.md.
Boundary markers: The instructions do not specify explicit delimiters or "ignore embedded instructions" warnings when interpolating external content into the agent's creative synthesis process.
Capability inventory: The skill possesses the capability to execute renku CLI commands (apply, import, set-placement) which persist data to the project state, and basic shell commands for file system navigation within project-relative paths.
Sanitization: The skill mitigates risks by requiring a mandatory validation step (renku lookbook validate) before any generated JSON is applied to the project workspace.

lookbook-designer