The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of CLI tools via uv run and uv tool. This includes generation commands (comfy-imagegen), gallery management (comfy-media), and setup utilities (comfy-tools-setup). These are standard execution patterns for local AI agent skills.
[EXTERNAL_DOWNLOADS]: The skill utilizes a dedicated utility, comfy-model-downloader, to fetch AI model weights (checkpoints, LoRAs, and VAEs) to /mnt/models/comfyui. It also references remote API services for Grok Imagine and Atlas Cloud, which are documented features for non-local generation.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided text prompts and input images through CLI arguments. While this represents an attack surface for indirect prompt injection, the skill uses standard quoting and parameterization for the comfy-imagegen commands.
Ingestion points: User-provided prompts and image paths passed via --prompt and --input flags in SKILL.md.
Boundary markers: None explicitly defined in the instruction text; however, the CLI tools use standard shell argument delimiters.
Capability inventory: The skill has the ability to execute shell commands (uv run), write output files to the outputs/ directory, and perform network requests to specific AI provider APIs.
Sanitization: The skill instructions do not specify sanitization logic, relying on the underlying comfy-imagegen CLI to handle string inputs safely.

comfy-imagegen