This project provides a standalone viewer for extracting and managing Stable Diffusion image generation prompts, catering to users who need to access this metadata outside of web UIs. It offers both GUI and CLI interfaces for reading, editing, and exporting prompts from various image formats.

How It Works

The tool parses image metadata, specifically looking for embedded prompt information generated by different Stable Diffusion interfaces. It supports multiple formats, including PNG, JPEG, and WEBP, and can detect the generation tool. The application allows users to easily copy prompts, remove them from images, or export them to text files.

Quick Start & Requirements

• Installation:
  • Windows/macOS: Download executable from GitHub Releases.
  • macOS (Homebrew): brew install --no-quarantine receyuki/sd-prompt-reader/sd-prompt-reader
  • Linux/Pip: pip install sd-prompt-reader or pipx install sd-prompt-reader
• Prerequisites: Python 3.10+ (for pip install), tkinter package (for Linux).
• Usage: Launch GUI via sd-prompt-reader or CLI via sd-prompt-reader-cli.
• Docs: README

Highlighted Details

• Supports a wide range of Stable Diffusion UIs including A1111, Easy Diffusion, StableSwarmUI, Fooocus-MRE, NovelAI, InvokeAI, ComfyUI, and Draw Things.
• Offers features like prompt copying to clipboard, removal from images, and export to text files.
• Includes a ComfyUI node for seamless integration into workflows.
• Provides both GUI and command-line interfaces for flexibility.

Maintenance & Community

The project is actively maintained by receyuki. Further community engagement details are not explicitly listed in the README.

Licensing & Compatibility

The project appears to be available under a permissive license, though not explicitly stated in the README. Compatibility for commercial use is likely, but a definitive license statement is absent.

Limitations & Caveats

The README notes potential issues with ComfyUI when custom nodes are used or workflows are overly complex, as metadata reading might fail. Editing images converts them to A1111 format. macOS users may encounter quarantine issues with unsigned executables. A false positive malware alert from antivirus software due to pyinstaller is also mentioned.