Dataset Tools is Back!
https://github.com/Ktiseos-Nyx/Dataset-Tools
We'll provide the source code HERE but if you want the Executable, you'll need to head to github for security reasons. Windows and Linux users get executables, and MacOS etc have to use command line.
Current Demo: https://dataset-tools-three.vercel.app
About
Dataset Tools NextJS Edition is a local-first web application for viewing EXIF metadata from AI-generated images and camera photos, including the software used to create AI images and the camera details embedded in photos.
Installation
Production mode; dev mode differs — see Development Mode.
Requires Node.js 20.9+ to run the application. Development installs want Node.js 22+ — a dev-only dependency (concurrently) requires it, and npm only warns (not fails) on older versions.
git clone https://github.com/Ktiseos-Nyx/Dataset-Tools.git
cd Dataset-Tools
npm install
npm run build
npm startOpen http://localhost:3000.
Desktop App (Windows)
A standalone .exe lives on the Releases page — no Node.js, no browser, no npm. It bundles Next.js and runs the same metadata engine as the web build, but against any folder on your machine. This comes in TWO flavors: Portable and NSIS based installer.
Portable works great for travel, or lower end machines that don't want to fuss with the full installation.
Smart Screen Issues
"Windows protected your PC"?
On first launch Windows will probably throw up a SmartScreen warning. That's not malware — it's because the exe isn't signed with a commercial code-signing certificate, which costs a few hundred dollars a year. Same thing every unsigned open-source desktop app deals with.
To run it:
Click More info
Click Run anyway
Two ways to check it yourself before running:
Every release ships a VirusTotal scan link so you can see the file checked against 70+ antivirus engines. You'll occasionally see 1–2 "generic heuristic" flags on a fresh Electron build — that's normal for unsigned apps, not a real detection.
The whole app is open source — the source and the build configuration are all here, so you can inspect exactly how the exe is produced.
But I hate SmartScreen!
A portable windows exe with no requirements to install is now available on the most recent releases.
Future versions for Linux are coming
This portable edition provides no requirements (as far as we're aware) and can be more useful than installing. It keeps its settings and thumbnail cache in a
datafolder created next to the EXE. API keys in that folder are stored scrambled — unreadable at a glance, but not encrypted — so if you run the EXE from a folder other accounts on the machine can access, assume those users can read them.
Usage
Start the app:
npm start→ openhttp://localhost:3000Browse files: Use the file tree sidebar, or click the folder icon to pick any directory.
Drag & drop: Drop an image anywhere in the app — it'll find the folder and load thumbnails.
Inspect metadata: Click any image → metadata panel shows prompts, parameters, LoRAs, and workflow info. Safetensors files show model architecture, training details, and dataset tags.
Edit metadata: Click the edit button on any text field to modify prompts and parameters directly in PNG files.
Customize: Settings panel has theme, accent colors, font size, thumbnail size, and file display options.
Note on folder browsing: The desktop app uses a native folder picker; the web build uses browser-local folder access (File System Access API). Subfolder navigation and remembering the folder between visits are still in progress — track issue #220.
When metadata fails to parse
Check browser console for parser logs.
Note the workflow structure (ComfyUI? A1111? Custom nodes?).
File an issue with:
Console error snippet
Workflow type + custom nodes used
Minimal repro image (if shareable)
Your OS if possible.
Version of the application.
If you're running the legacy Python edition, it is no longer supported — file issues against the Next.js edition only.
Community-Driven Development
This project is inspired by stable-diffusion-prompt-reader and thrives on community contributions. Found a bug? Have a workflow that won't parse? Want to add support for a new tool? We welcome forks, fixes, and pull requests!
Found a parsing failure?
Open an issue with the details above. ComfyUI always has edge cases, and while we intend to catch almost everything we can with the lookup on desktop and web, there are always new software tags and nodes.
Want to improve the parser?
Fork repo →
npm install→npm run devMetadata extraction lives in
app/api/metadata/route.tsTest with your own images or AI-generated samples
Submit a PR with before/after evidence
Description
Version release notes are here: https://github.com/Ktiseos-Nyx/Dataset-Tools/releases/tag/v0.96.0



