The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute the runcomfy CLI tool to perform image edits based on user intent. This requires passing user-defined parameters into shell-executed commands.\n- [EXTERNAL_DOWNLOADS]: The skill requires installation of the @runcomfy/cli Node.js package. This is the official client for the RunComfy service mentioned in the documentation.\n- [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted user input that could influence the behavior of backend models.\n
Ingestion points: User-supplied prompt strings and image_urls / mask_image URLs are used as inputs for model execution.\n
Boundary markers: The instructions do not specify any delimiters to isolate user-supplied prompts from internal instructions when building the JSON payload.\n
Capability inventory: The skill uses the runcomfy CLI, which has the capability to download files to a local directory and communicate with external APIs.\n
Sanitization: The skill does not describe any methods for validating or sanitizing user-provided text or URLs before processing.

image-edit