The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill specifies the installation of the sharp package via npm in its package.json to perform image manipulation tasks.
[COMMAND_EXECUTION]: The skill includes several utility scripts (generate_image.js, resize_cover.js, remove_bg_enhanced.js) which are intended to be executed by the agent to generate and process image files locally.
[DATA_EXFILTRATION]: The script scripts/generate_image.js makes network requests to https://api.apiyi.com to send text prompts and receive image data. This is a functional requirement for its image generation capabilities and does not involve the exfiltration of sensitive local data or credentials.
[INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection as it interpolates user-provided text prompts into API requests for image generation.
Ingestion points: User prompts are passed as arguments to the generate_image.js script.
Boundary markers: No specific delimiters or safety warnings are included to wrap the user-provided prompt data.
Capability inventory: The skill scripts have the capability to perform network requests (https.request) and write files to the local disk (fs.writeFileSync).
Sanitization: No sanitization or filtering of the input prompt is performed before it is sent to the image generation API. The impact is limited because the resulting output is a static image file.

wechat-red-envelope-cover-designer