Crop JPEG/PNG photos from natural language prompts using GPT-5 vision analysis.
npx skills add https://github.com/ppx123-web/claude-config --skill photo-clipper
Please help me install this Agent Skill. Skill: photo-clipper Source: https://github.com/ppx123-web/claude-config/tree/main/skills/photo-clipper Command: npx skills add https://github.com/ppx123-web/claude-config --skill photo-clipper
This Skill automates the time-consuming task of cropping photos according to natural language prompts, leveraging GPT-5 vision analysis to produce well-composed results without manual editing.
Provide a photo and a natural language crop prompt to clip accordingly.
High-intent search queries and answers about installing and using this skill.
You can crop a photo by providing the image and a natural language prompt describing the desired framing. The tool applies GPT-5 vision analysis to identify subject emphasis and generate balanced crops.
Yes, intelligent cropping supports both JPEG and PNG input files. The system includes input validation to ensure compatibility and processes the provided photo to generate the requested crop.
Using AI-assisted vision analysis identifies framing improvements and edge removals while preserving subject focus. It enforces a safe limit of 50% removal per dimension to prevent over-cropping and maintain composition.
To crop photos non-destructively, the system always saves the result to a new file with a "-clipped" suffix. This ensures your original image remains untouched while the tool provides the reasoning for the crop.
Yes, the main limitation for image cropping is a strict safety check that enforces a maximum removal of 50% per dimension. This prevents excessive cropping and ensures the output remains a balanced photo.
Your photo clip may fail due to unsupported file formats; only JPEG and PNG inputs are accepted. The system provides clear error handling to indicate validation issues or if the requested crop exceeds safe limits.
hasnaExtract text and structured data from images and PDFs into plain text, Markdown, or JSON.
m0n0x41dFrame problems and guide structured reasoning for architecture decisions.
Buntyarora97Automate image resizing, cropping, filtering, and format optimization.
jonnabioIntegrate vision-language models for visual question answering and image captioning.
