isGenAI.com

A working example, with the files

Extract a ComfyUI workflow from a PNG

An original ComfyUI PNG may contain an editable workflow and a separate API graph. Inspect those records directly when you need the workflow structure rather than just one text prompt.

By isGenAI · Published

Open the workflow viewerTry the example files

Follow the steps

  1. Download the PNG with workflow records below, then choose it in the workflow viewer.
  2. Read the two record types. Each controlled example contains two nodes: LoadImage and SaveImage.
  3. Open the saved JSON, inspect its values and download the record you need. Keep the two files labelled by their record type.

What this example contains

We attached a small hand-built graph to code-drawn pixels. Both records describe a LoadImage → SaveImage path. They demonstrate graph extraction; we did not run the graph to create this picture.

The prompt text extractor returns no saved generation prompt for this file. That is expected: this simple example has no text-conditioning generation path. A useful workflow and a recoverable positive prompt are different results.

Inspect before opening elsewhere

The viewer reads JSON as data. It does not install custom nodes, download referenced files or run the graph. The saved JSON may contain paths or names supplied by its author.

A downloaded workflow does not include model weights or input assets. Verify dependencies in your own environment before trying to run any workflow.

If the PNG contains no workflow

Try the original export rather than a re-encoded preview. A missing record, an unsupported structure or the reader’s limits can prevent recovery. This viewer supports PNG and standalone JSON; other media formats require a suitable reader.

Download · Run · Compare

Try the verified example

We drew the picture by code and attached controlled records. The saved prompts and model names are demonstration data. These files test record reading; no image generator, screenshot app or social network was used to create these test results.

Download editable workflow JSON · Download API graph JSON

Observed output
[
  {
    "field": "workflow",
    "nodes": 2,
    "node_types": [
      "LoadImage",
      "SaveImage"
    ]
  },
  {
    "field": "prompt",
    "nodes": 2,
    "node_types": [
      "LoadImage",
      "SaveImage"
    ]
  }
]
Hashes and reproduction
  • workflow-records.png · 14,013 bytes427aa019b768aaaeb2fb59724a4e49772e041ee6091f47cffc7ec6a35867dc08
  • without-records.png · 13,179 bytesacdc3a0a5e011cf57af9bf0f1da74c75c2d280a2cc9af9297410c529f86932cf

Fixtures and observations are CC0-1.0. Download all observations and versions or read the procedure. The complete Node.js reproduction script (requires sharp) creates the same test inputs and compares bytes and pixels. The isGenAI source checkout additionally verifies saved fields with npx tsx scripts/build-guide-lab.ts.

For the separate export study, explore the 18 metadata-retention cases.

Sources and method

The linked primary sources describe the formats. The downloadable fixture outputs were checked with automated assertions against the current isGenAI reader. These guides were prepared with AI assistance and reviewed against those examples.

Report an error or unsupported example. For a signed record, inspect Content Credentials.