The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the dlazy CLI tool using node:child_process.spawn within scripts/lib/providers.mjs. The specific binary path can be overridden by the DLAZY_BIN environment variable.\n- [EXTERNAL_DOWNLOADS]: The scripts/lib/providers.mjs script makes outbound requests to several well-known AI service providers, including OpenAI, Google Gemini, Fal.ai, and Replicate, to process image tasks.\n- [DYNAMIC_EXECUTION]: scripts/gen.mjs dynamically constructs command-line arguments at runtime based on task configurations and brand guidelines before passing them to the execution logic.\n- [INDIRECT_PROMPT_INJECTION]: The skill incorporates user-provided prompts and external brand configuration into its instructions for AI models, creating a surface for indirect injection.\n
Ingestion points: User-supplied prompts via --prompt or --prompt-file in scripts/gen.mjs, and brand configuration data from YAML files in scripts/brand.mjs.\n
Boundary markers: None identified; input is directly concatenated or interpolated into the final prompt string.\n
Capability inventory: The skill possesses capabilities for subprocess execution (spawn), network communication (fetch), and file system writes (writeFile).\n
Sanitization: There is no evidence of input validation or escaping for the prompt strings before they are transmitted to external AI providers.

item-repair