The Agent Skills Directory

[COMMAND_EXECUTION]: The skill facilitates the execution of system commands and local scripts to manage ComfyUI.
Evidence: Instructions call python scripts/comfyui_cli_bridge.py and the comfy CLI tool for operations like launch, run, and models-search.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of ComfyUI workflows, which creates an attack surface for indirect prompt injection.
Ingestion points: The skill reads and converts workflow data from WORKFLOW.json as seen in SKILL.md.
Boundary markers: The instructions recommend using --json for structured data and performing discovery/validation before environment changes.
Capability inventory: The skill can execute commands, manage background processes, and install/update dependencies using the comfy binary and bridge scripts.
Sanitization: No specific input sanitization steps for the JSON workflow files or conversion process are described in the instructions.

comfyui-cli