diff --git a/README.md b/README.md
index b460849..1c9fa5f 100644
--- a/README.md
+++ b/README.md
@@ -23,7 +23,7 @@ A self-hosted web app for organizing your 3D-printing files (STL, 3MF, OBJ, FBX,
 - **Notes** — free-form notes per model, saved as you type.
 - **Search, filters & sorting** — search across names, descriptions, notes, tags, file names and designers; filter by format, source and license; include *or exclude* tags (match any/all); sort and group by source or format. Press <kbd>/</kbd> to search.
 - **Duplicate finder** — every file gets a SHA-256 content hash plus (for STL/OBJ/3MF) a geometry fingerprint, computed in the background. Models sharing files are grouped as *exact copies* or *same shape* (renamed, re-exported, converted — or mirrored, so check those), with a recommended copy to keep. Keeping one merges the others' tags, collections and notes into it and moves them to the trash; groups can be dismissed as "not duplicates".
-- **Watch a MakerWorld collection** — paste a collection link (private ones work, using your MakerWorld token) and Printhive checks it on a schedule you choose, importing models added to it into a collection of the same name here, optionally AI-tagging the new ones. Each check reads the whole collection (paging through MakerWorld's own load-more call, up to ~640 models) and imports up to 15 of the new ones, spaced out, returning shortly after if more are waiting; models you remove on MakerWorld stay in your library. Each watch shows when it last ran, what it imported, and why a check failed — with a one-click token update when MakerWorld refuses your token. A check imports each model's page — name, description, picture, licence — but not its files: MakerWorld only lets an account download a couple in a row, so models arrive listed as still without files and you fetch the ones you want with "Get files".
+- **Watch a MakerWorld collection** — paste a collection link (private ones work, using your MakerWorld token) and Printhive checks it on a schedule you choose, importing models added to it into a collection of the same name here, optionally AI-tagging the new ones. Each check reads the whole collection (paging through MakerWorld's own load-more call, up to ~640 models) and imports up to 15 of the new ones, spaced out, returning shortly after if more are waiting; models you remove on MakerWorld stay in your library. Each watch shows when it last ran, what it imported, and why a check failed — with a one-click token update when MakerWorld refuses your token. A check imports each model's page — name, description, picture, licence — but not its files, so a large collection arrives quickly; models are listed as still without files and you fetch the ones you want with "Get files". MakerWorld refuses downloads fired off in quick succession (a few seconds apart, as a check would), but spacing them out is fine — 67 models in a row downloaded about 20 seconds apart without a single refusal.
 - **AI tags & descriptions** — bring your own key for DeepSeek (the default), OpenRouter, OpenAI or any OpenAI-compatible server (Ollama, LM Studio). Tag the whole library, only models not yet tagged, or a single model from its page. Each model's name, file names and (optionally) thumbnail are sent; tags are only ever added and descriptions only filled where empty unless you ask. Optionally it also gives models with unclear names (file-name leftovers, typos, site boilerplate) a clearer one; the old name is kept and can be restored with Undo on the model page. It can also suggest collections (reusing yours where they fit): suggestions wait on the AI page, grouped by name, until you create/add or dismiss them. Each run can do any mix of tags & descriptions, names and collections, and a reply that isn't the JSON asked for is retried once. Runs show live progress, tokens and cost, can be stopped and resumed, stop at an optional spending limit, and pause (rather than keep spending) if the server restarts. "Save & test connection" in Settings checks the key, model and whether it accepts images.
 - **Search by meaning** — switch the search box to *Meaning* to find models by what they are ("wall hook" finds a "coat hanger bracket"). Uses a small embedding model (bge-small-en-v1.5, ~34 MB) that runs on the server, so it's free and private; the index updates in the background as models change.
 - **Automatic tags** — strict whole-word rules over names, file names and archive contents (part types, mechanisms, materials, techniques) plus your own "word → tag" rules. Switch it off (optionally removing its tags) or re-apply it any time. Tags you add yourself are never changed; AI and rule tags are marked on the model page.