The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the infsh command-line tool to run remote image processing applications (e.g., Real-ESRGAN, Topaz). These executions are restricted to the inference.sh ecosystem and are required for the skill's core functionality.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it processes untrusted external data in the form of image URLs.
Ingestion points: The image_url parameter used in infsh app run commands within SKILL.md.
Boundary markers: Absent; the URL is passed directly as part of a JSON input string.
Capability inventory: The skill has the capability to execute bash commands via the infsh tool as defined in the allowed-tools metadata.
Sanitization: No specific sanitization or validation of the URL or the image content is performed within the skill definition, relying instead on the remote application's internal safeguards.

image-upscaling