The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documentation includes shell commands for directory navigation, setting environment variables, and running Python scripts to perform image generation.
[INDIRECT_PROMPT_INJECTION]: Untrusted user-provided articles and content are directly interpolated into image generation prompts without the use of boundary markers or sanitization logic. * Ingestion points: User-provided article or content text processed during the generation flow. * Boundary markers: None identified in the prompt templates to distinguish user content from instructions. * Capability inventory: Shell command execution and network API requests through the referred generation script. * Sanitization: No input filtering or validation is performed.
[DYNAMIC_EXECUTION]: The skill provides a command that uses a Python one-liner ('python3 -c') to dynamically parse and extract API keys from a JSON configuration file at runtime.

hd-infoimage