The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from screenplays, character DNA files, and location anchors to generate storyboard panels. This ingestion of external data represents a potential attack surface where malicious instructions could be embedded to influence the agent's output. However, the skill's impact is restricted to generating text-based storyboard data and prompts within the project's local filesystem.
Ingestion points: Files located under project/screenplay/, project/continuity/, project/production-design/, and project/characters/ as described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or validation rules to separate screenplay data from agent instructions.
Capability inventory: The skill is limited to writing storyboard descriptions, prompts, and plans to project/storyboards/. It does not possess network access or the ability to execute arbitrary system commands.
Sanitization: No explicit sanitization or filtering of the ingested screenplay data is documented.
[SAFE]: The skill follows established platform patterns for AI production workflows, including the use of 'locked anchors' for consistency and coordination with other specialized skills in a pipeline. All file operations are constrained to the local project directory.

creator-storyboard-artist