The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses child_process.spawn in scripts/lib/providers.mjs to invoke the dlazy command-line utility for image processing.
[EXTERNAL_DOWNLOADS]: The gen.mjs script interacts with external APIs from trusted providers like OpenAI, Google, and Fal, and fetches product images from remote URLs specified by the user.
[DYNAMIC_EXECUTION]: The skill dynamically generates CLI command strings based on task definitions. It includes appropriate escaping logic to mitigate command injection risks.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user input and images, creating a potential surface for indirect prompt injection. \n
Ingestion points: User prompts and images ingested via the gen.mjs entry point in skill.md. \n
Boundary markers: No explicit delimiters are used to isolate untrusted data within the prompt templates. \n
Capability inventory: The skill can execute local CLI tools, write generated images to the file system, and access remote AI endpoints. \n
Sanitization: Input content is not sanitized against LLM-specific injection attacks before interpolation into prompts.

wear-everything