The Agent Skills Directory

[SAFE]: No security issues were identified in the skill's instructions or reference documents. The skill follows best practices for structured text generation and focuses on authoring video-related design documents.
[NO_CODE]: The skill does not contain any executable scripts, binary assets, or external library dependencies, reducing the direct execution attack surface.
[PROMPT_INJECTION]: The skill utilizes a structured workflow and operational boundaries to define the agent's role as a script designer, avoiding any patterns designed to bypass safety filters or reveal system prompts.
[INDIRECT_PROMPT_INJECTION]: The skill provides an interface for processing untrusted user briefs into video scripts, which serves as a potential vector for data-driven influence.
Ingestion points: User-provided briefs and requirements are ingested during the BRIEF phase.
Boundary markers: A multi-phase workflow is used to ground the agent's behavior, though explicit delimiters for untrusted input are absent.
Capability inventory: The generated scripts are intended for use by the 'Director' (Playwright) and 'Reel' (CLI) agents, which possess executable capabilities.
Sanitization: No explicit sanitization of user-provided content is mentioned in the workflow.

cue