The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes child_process.spawn within scripts/lib/providers.mjs to execute the dlazy CLI tool. This is the primary mechanism for invoking image generation models. The script implements argument escaping to mitigate command injection risks.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to interface with the vendor's infrastructure at api.dlazy.com and files.dlazy.com. It also communicates with well-known AI service providers including OpenAI, Google Gemini, Fal.ai, and Replicate. Additionally, scripts/gen.mjs may download image assets from user-provided URLs as part of the generation workflow.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from brand.yaml files and user-provided prompts, which are interpolated into the final instructions sent to AI models.
Ingestion points: scripts/brand.mjs reads brand configuration files, and scripts/gen.mjs accepts user-defined prompts.
Boundary markers: Prompts are structured using predefined templates in skill.md and scripts/brand.mjs.
Capability inventory: The skill possesses capabilities for network requests, shell command execution via spawn, and local file writing for output assets.
Sanitization: Shell arguments are escaped in scripts/lib/providers.mjs, though raw prompt text is passed to the AI models as is typical for generation tools.

flat-lay