SCRUB
0 bytes uploaded — 100% in-memory ArrayBufferOnline — still 100% localSession bytes saved0 B

Remove AI Generation Metadata & Prompts (Midjourney, Stable Diffusion, ComfyUI)

AI image tools often embed the full generation prompt, model name and seed in the file. Inspect those markers and remove them before you share the image.

Drop images here

JPEG, PNG or WebP · up to 25 MB each · add as many as you like

Try a sample

Binary telemetry

  1. [1/4] ReadFile → ArrayBuffer
  2. [2/4] ParseWalk container segments
  3. [3/4] StripDrop selected metadata
  4. [4/4] RebuiltEmit cleaned bytes

Cleaning profile

Only the categories you check below are stripped.

Write a fresh Make, Model and ISO block into every cleaned image.

No images queued

Add JPEG, PNG or WebP images above to build a batch. You can queue 30+ files and clean them all at once.

Single-image inspector

nothing selected

No image loaded

Add an image above to inspect its metadata in detail.

Awaiting an image

Drop a JPEG, PNG or WebP above to read its metadata. Parsed fields appear here as a grouped evidence table.

0 B uploaded · 0 bytes sent

How It Works

  1. 1

    Drop an AI-generated image; the parser scans EXIF, XMP and PNG text chunks for generation markers.

  2. 2

    The inspector surfaces the prompt, model and workflow fields it finds, and flags any C2PA AI assertions.

  3. 3

    AI markers and C2PA provenance are pre-selected so both are removed.

  4. 4

    Clean and download an image with no generation metadata attached.

Binary Format Breakdown

Container segments this tool inspects and how each is handled
SegmentCarriesAction
PNG tEXt / iTXtComfyUI workflow JSON, Automatic1111 parameters, promptsWhole text chunks dropped
XMPMidjourney job IDs, model names, generation parametersRewritten in place to drop AI fields
C2PA / JUMBFSigned AI-generation assertionsManifest segments removed

Frequently Asked Questions

Which AI tools embed metadata?

Midjourney, Stable Diffusion (Automatic1111 and ComfyUI), Adobe Firefly and many others write prompts, model names, seeds and workflow graphs into the file.

Is the prompt really stored in the image?

Often yes. ComfyUI stores its entire workflow graph and Automatic1111 stores the full prompt and settings, usually in PNG text chunks or XMP.

Does removing AI metadata change the picture?

No. Only the metadata segments are removed; the image pixels are copied verbatim.