Color Space Analyzer

ColorSpaceAnalyzer is a diagnostic node with zero knobs: drop an image in (tensor or file path), and it returns three JSON strings describing the image's color state - color_space_info, color_stats, and recommendations. Think of it as the pack's triage node. It won't fix anything; it tells you what's off so you know which fix node to reach for.

How it works

It computes per-channel statistics on the image (mean, range, per-channel balance), then runs a handful of heuristics that produce the recommendations. The heuristics in the source are telling: values exceeding 1.0 get flagged for normalization, negative values trigger a "check your color space" warning, a mean difference of more than 0.1 between R and G or G and B flags unbalanced channels for color correction, and it converts to HSV to measure mean saturation - low saturation (< 0.3) suggests a boost, high (> 0.8) suggests a cut.

So the practical reading: color_stats is the raw data, and recommendations is the pack's own opinionated summary of what to do next. When it says "Color channels are unbalanced, consider color correction," that's the pack telling you to reach for BrightnessContrast or ColorBalance. When it says "Low saturation detected," that's Saturation. It's the pack's internal knowledge surfaced as a decision aid.

The inputs that matter

Just input_mode (tensor or file). Everything else is automatic.

• color_space_info - what the node thinks about the image's space and value range.
• color_stats - the per-channel numbers, for your own analysis.
• recommendations - the actionable list.

Where it fits

This is a "run it on a reference image, then on your output, and diff" node. Feed it the same image before and after a conversion and the recommendations tell you whether the conversion fixed or worsened the imbalance. It's also a decent onboarding tool: if you're new to color grading and don't know where to start, run the analyzer and it points at the specific problem - dark, desaturated, unbalanced - instead of leaving you to guess.

Install

Same pack, same drill:

cd ComfyUI/custom_nodes
git clone https://github.com/APZmedia/ComfyUI-color-tools

Or ComfyUI Manager → search "ComfyUI Color Profile Reader" → Install → restart. Mind the placeholder clone URL in the README (yourusername/... - it's APZmedia/ComfyUI-color-tools). Needs OpenCV for the HSV saturation check (optional dep, installed by the startup install.py). And be honest about its limits: these are statistical heuristics, not perception. An intentionally desaturated, moody image will get flagged "low saturation" and told to fix a problem it doesn't have. Treat the recommendations as hints with context, not commands.