Ino Json Get Field

You know the drill: some node hands you a JSON string - an LLM's answer, a model config, an S3 config - and what you actually need is the one value buried inside it. Ino Json Get Field is the pair of tweezers. You give it the JSON and a field name, and it pulls that value out as a real typed output you can wire straight into the next node.

It's part of ComfyUI Ino Nodes (nobandegani/ComfyUI-InoNodes), a 125+-node pack built entirely on the modern V3 schema. This is one of its JSON helpers, and honestly it's the one you'll reach for most: a huge chunk of this pack's own design (model configs, LoRA configs, S3 configs) is JSON strings flowing between nodes, so being able to rip a single field out is what makes those chains readable instead of mysterious.

How it works

Under the hood it's embarrassingly simple, which is the point. The node parses base_json through the pack's shared inopyutils JSON helper, then indexes into the result by field_name. That's it. What it gives you back is the value itself, not a re-serialized string - so if the field holds a number, a boolean, or a nested object, that's what comes out the field_value output. For a beginner this matters more than it sounds: it means you can chain Ino Json Get Field into another node that expects a typed value and ComfyUI won't complain about types.

The inputs and outputs that matter

Only two inputs, both strings:

• base_json - the JSON text to read from.
• field_name - the key to extract.

Three outputs, and here's the pattern you'll see all over this pack: success (boolean), message (string), and the thing you actually wanted, field_value (any type). The success/message pair is the pack's way of doing error handling in the graph - every node in Ino Nodes speaks this same language, so once you learn it here you've learned it everywhere.

Common issues

The failure modes are friendly because they're explicit. If base_json isn't valid JSON, or the field doesn't exist, you get success: False and a message telling you what broke - and field_value comes back empty rather than crashing the whole workflow. That's the difference between a node designed for automation and a quick hack. Two things to watch: field names are case-sensitive and exact, and if you're pulling a field that's nested inside an object, this node only does top-level keys. For deeper paths you'd reach for the pack's Ino Json Set Field / JSON manipulation friends, or preprocess the JSON upstream.

Installing it

ComfyUI Ino Nodes installs like any custom node pack:

• ComfyUI Manager (easiest): search for "ComfyUI Ino Nodes" and hit install, then restart ComfyUI.
• Manual:

  cd ComfyUI/custom_nodes
  git clone https://github.com/nobandegani/ComfyUI-InoNodes.git
  cd ComfyUI-InoNodes
  pip install -r requirements.txt

  Then restart ComfyUI.

One thing to know before you install: the pack requires inopyutils (its shared utility library) and targets the newer V3 schema, so it wants a reasonably current ComfyUI. It's also a niche pack - the 0-impression search volume on this page is not a fluke; there's basically no community chatter about it, which makes the README's documentation your best friend. But for pulling fields out of JSON mid-workflow, it does exactly what it says on the tin.