protecting your timeline from the egregore since 2026
Some people find that a significant percentage of their timeline consists of accounts using aesthetically identical chibi avatars posting aesthetically identical content. This extension addresses that.
A bundled ONNX classifier scans avatars as you scroll. When it spots a match, you pick what happens:
- Hide — collapsed behind a click-to-reveal row.
- Fade — visible but at half opacity.
- Debug — borders and confidence scores on every post.
- Off — does nothing.
The popup tracks session stats (posts scanned, match rate, last sighting), keeps a list of detected accounts you can exempt individually, and collects avatar data you can export for offline labeling.
Everything runs locally. No server calls, no telemetry, nothing leaves your browser unless you explicitly export it.
There is no Chrome Web Store release. Install from GitHub Releases instead:
- Download the latest
milady-shrinkifier-vX.Y.Z-unpacked.zipfrom the Releases page. - Unzip it somewhere permanent on disk.
- Open
chrome://extensions. - Enable
Developer mode. - Click
Load unpacked. - Select the unzipped folder.
All scores below come from the current manually reviewed exported corpus.
- Precision — when the extension filters a post, how often it's right.
- Recall — of the Milady-style avatars in the evaluation set, how many it catches.
- Evaluation corpus —
7,695exported avatars (437milady,7,258not_milady). - This is broader than the blind split and still fully human-labeled, but it is not a blind benchmark.
| Version | Run | Training mix | Precision | Recall |
|---|---|---|---|---|
v0.2.2 |
20260327T142224Z |
Milady Maker + 2,596 manually tagged avatars |
0.9961 |
0.5904 |
v0.3.0 |
20260327T212453Z |
+ Remilio, Pixelady + 2,967 manually tagged avatars |
1.0000 |
0.7208 |
v0.4.0 |
20260328T144735Z |
+ 5,715 manually tagged avatars |
0.9971 |
0.7918 |
v0.5.0 |
20260328T223931Z |
+ 6,773 manually tagged avatars |
0.9952 |
0.9451 |
v0.6.0 |
20260329T124912Z |
+ 7,370 human-reviewed avatars |
0.9952 |
0.9474 |
v0.7.0 |
20260329T181946Z |
+ 7,695 human-reviewed avatars |
0.9951 |
0.9291 |
All rows were re-evaluated on the same manually reviewed exported corpus on March 29, 2026, so they are directly comparable.
See DEVELOPMENT.md for build commands and debugging.
For a user-facing walkthrough of how the offline training loop works end to end, see docs/training-pipeline.md.
