Normalize input images to 896×896 RGB PNGs for local vision models.
npx skills add https://github.com/BaiGanio/aperio --skill preprocess-image
Please help me install this Agent Skill. Skill: preprocess-image Source: https://github.com/BaiGanio/aperio/tree/main/skills/preprocess-image Command: npx skills add https://github.com/BaiGanio/aperio --skill preprocess-image
Local vision models often require input images in a standard RGB PNG format with fixed dimensions. This skill ensures any input image is converted to a consistent 896×896 RGB PNG, removing alpha channels, converting color spaces as needed, and filling transparent areas with a solid background so the model receives a predictable input.
Call preprocess_image with a local image path (or base64 data) to obtain a normalized 896×896 RGB PNG suitable for local vision models.
High-intent search queries and answers about installing and using this skill.
To normalize images for local vision models, convert them to a consistent 896×896 RGB PNG by removing alpha channels, converting color spaces, and applying letterbox resizing to prevent format-related inference errors.
Yes, you can use sharp to remove alpha channels and convert color spaces, producing a ready-to-use base64 RGB PNG specifically formatted for local vision model ingestion.
The best way to preprocess transparent PNGs is to fill alpha areas with a solid white or dark background while resizing the image to a fixed square dimension like 896, 512, or 1024 pixels.
Preprocess-image accepts both local file paths and base64 data inputs, returning a normalized base64-encoded PNG alongside a confirmation message for downstream agents.
Local vision models require fixed square dimensions like 896x896 RGB PNGs to receive predictable inputs, avoiding processing errors caused by varying channels, formats, or dimensions from web uploads and local files.
lgoodcodeAutomate image upload processing, moderation, and CDN delivery with Sharp and Gemini.
hotriluanProcess multimedia inputs for multimodal analysis, transcription, captioning, and generation.
DFly7Integrate Core ML models into iOS apps for on-device inference.
mgd34msuResize, convert, and optimize images with the Sharp Node.js library.
