The Agent Skills Directory

[SAFE]: The skill does not contain any malicious code, hidden URLs, or unauthorized network operations. It strictly follows a creative workflow for generating text-based prompts for image generation tools.- [INDIRECT_PROMPT_INJECTION]: The skill processes external screenplay data, creating a potential surface for indirect prompt injection. However, this risk is mitigated by the skill's lack of dangerous capabilities and its reliance on highly structured output templates that enforce strict formatting rules. Ingestion points: The agent processes user-provided drama scripts and summaries as described in Step 1 of the SKILL.md workflow. Boundary markers: The instructions employ rigid formatting requirements (e.g., specific 4-view grids for characters and props) and explicit asset numbering systems that constrain the agent's output. Capability inventory: The skill is limited to generating text prompts and referencing project-internal documentation; it does not execute system commands, access sensitive local files, or perform network operations. Sanitization: The workflow includes a mandatory 'Step 5' consistency check to validate output against the project's internal asset ledger.

sd-asset-engine