The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill translates natural language user descriptions into structured prompts for image generation models, which represents an attack surface for indirect prompt injection. 1. Ingestion points: User visual requirements are processed to build prompts. 2. Boundary markers: No specific delimiters or safety instructions are defined to encapsulate user input. 3. Capability inventory: The workflow documentation mentions writing generated images to a local assets directory and making external API calls to image generation backends. 4. Sanitization: No input validation or sanitization logic is present in the documentation.
[NO_CODE]: The skill consists entirely of markdown documentation and instructional guidelines with no scripts or executable components.
[SAFE]: Analysis of the guidelines and metadata found no evidence of malicious intent, obfuscation, or safety bypasses.

art-director