Anima Artist Probe Report

AnimaArtistProbe measures where your artists' styles live; this node makes that measurement useful. AnimaArtistProbeReport takes a probe id, waits until sampling is done, and renders the analysis: per-artist, per-layer influence as a bar chart, a contribution split with plain-language verdicts, per-step influence curves showing when each artist matters, and - the part that pays for itself - a suggested_chain you can wire straight back into AnimaArtistPack.

How to wire it

• probe_id - the STRING output of AnimaArtistProbe.
• trigger - optional, but recommended: connect any post-sampler output (e.g. the decoded IMAGE) so the report runs after sampling actually finished. Without a trigger you risk reading results before they exist.

What you get back

• report - the full analysis STRING: per-artist contribution split (dominant / balanced / weak / negligible), per-layer influence, and per-step influence curves showing when in sampling each artist matters.
• suggested_chain - since v27.6. Weights that equalize the measured per-artist influence, with layer routes preserved. Wire it into AnimaArtistPack.artist_chain and the pack auto-rebalances the mix toward what you actually measured instead of what you guessed.

Install

Manager → search "Anima Artist Mixer Forge", or:

cd ComfyUI/custom_nodes
git clone https://github.com/peter119lee/Anima-Artist-Mixer-Forge

then restart. No extra dependencies. Remove the original Anima-Artist-Mixer if installed.

My take

The suggested_chain output is the reason this node deserves a spot in your diagnostics workflow even though it "just" reports. The probe/report pair is the closest thing this pack has to an autotuner: run once, feed the suggestion back, and you've converted a dominated mix into a balanced one without hand-tuning four weights. The per-step curves are the underrated bit - they'll show you an artist that only matters in the first 20% of sampling, which is exactly the kind of finding you can't get from looking at final images.