feat(suggestions): tag-input dropdown searches the full prediction set

The typed dropdown sourced the threshold-filtered panel list (>= 0.70 general),
so low-confidence actions/features the model DID predict never appeared — forcing
hand-typed custom tags instead of accepting the model's canonical formatting.

Add a threshold override: SuggestionService.for_image(threshold_override=) and
GET /images/<id>/suggestions?min=<f> surface EVERY stored prediction (down to the
0.05 store floor), alias-resolved and normalized, still excluding applied/rejected
and unsurfaced categories. The suggestions store gains allByCategory + loadAll
(min=0); the dropdown searches that full set (cap 20), while the Suggestions panel
stays curated at the configured threshold. Accept/dismiss drop from both lists.

Operator-asked 2026-06-09. Test: a 0.30 general prediction is hidden by default
but surfaced with threshold_override=0.0; unsurfaced categories still excluded.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

backend/app/api/suggestions.py

@@ -11,8 +11,21 @@ suggestions_bp = Blueprint("suggestions", __name__, url_prefix="/api")
 
 @suggestions_bp.route("/images/<int:image_id>/suggestions", methods=["GET"])
 async def get_suggestions(image_id: int):
+    # ?min=<float> overrides the configured per-category thresholds so the typed
+    # tag-input dropdown can surface EVERY stored prediction (min=0), including
+    # low-confidence actions/features, in canonical formatting. Omitted → the
+    # curated above-threshold list the Suggestions panel uses.
+    override = None
+    raw_min = request.args.get("min")
+    if raw_min is not None:
+        try:
+            override = min(1.0, max(0.0, float(raw_min)))
+        except ValueError:
+            return jsonify({"error": "min must be a float in [0,1]"}), 400
     async with get_session() as session:
-        sl = await SuggestionService(session).for_image(image_id)
+        sl = await SuggestionService(session).for_image(
+            image_id, threshold_override=override
+        )
     return jsonify(
         {
             "by_category": {

backend/app/services/ml/suggestions.py

@@ -48,16 +48,33 @@ class SuggestionService:
             await self.session.execute(select(MLSettings).where(MLSettings.id == 1))
         ).scalar_one()
 
-    def _threshold_for(self, s: MLSettings, category: str) -> float:
+    def _threshold_for(
+        self, s: MLSettings, category: str, override: float | None = None,
+    ) -> float:
         # 'artist' (FC-2d-vii-c) and 'copyright' (2026-06-01) retired;
         # both fall through to the 1.01 "never surfaces" default like any
         # unsurfaced category.
+        # override (the typed-dropdown "show everything the model saw" mode)
+        # applies to the surfaced categories only — unsurfaced ones are already
+        # skipped before the threshold check, so they can't leak in.
+        if override is not None:
+            return override
         return {
             "character": s.suggestion_threshold_character,
             "general": s.suggestion_threshold_general,
         }.get(category, 1.01)
 
-    async def for_image(self, image_id: int) -> SuggestionList:
+    async def for_image(
+        self, image_id: int, *, threshold_override: float | None = None,
+    ) -> SuggestionList:
+        """Ranked suggestions for one image.
+
+        threshold_override surfaces EVERY stored tagger prediction (down to the
+        ingest STORE_FLOOR) regardless of the configured per-category suggestion
+        thresholds — backs the tag-input dropdown's "search all of the model's
+        predictions, including low-confidence ones, in the canonical formatting"
+        mode (operator-asked 2026-06-09). The Suggestions panel still calls with
+        no override so it stays the curated above-threshold list."""
         img = await self.session.get(ImageRecord, image_id)
         if img is None:
             return SuggestionList()
@@ -96,7 +113,7 @@ class SuggestionService:
             if category not in SURFACED_CATEGORIES:
                 continue
             conf = float(p.get("confidence", 0.0))
-            if conf < self._threshold_for(settings, category):
+            if conf < self._threshold_for(settings, category, threshold_override):
                 continue
             display = normalize_tag_name(name)
             if display is None:

frontend/src/components/modal/SuggestionsPanel.vue

@@ -62,7 +62,11 @@ const isEmpty = computed(() =>
   Object.values(store.byCategory).every(list => !list || list.length === 0)
 )
 
-watch(() => props.imageId, (id) => { if (id != null) store.load(id) }, { immediate: true })
+watch(() => props.imageId, (id) => {
+  if (id == null) return
+  store.load(id)        // panel: curated, ≥ threshold
+  store.loadAll(id)     // dropdown: full prediction set (low-confidence included)
+}, { immediate: true })
 
 // After a successful accept/alias-accept, refresh the modal's current
 // tag list so TagPanel's chip rail reflects the newly-attached tag.

frontend/src/components/modal/TagAutocomplete.vue

@@ -190,12 +190,17 @@ function scorePct (s) { return `${Math.round(s.score * 100)}%` }
 
 // This image's suggestions that match the typed query, minus any the server
 // autocomplete already returned (same name+kind) so a tag never shows twice.
+// Sources the FULL prediction set (allByCategory, down to the store floor) — NOT
+// the threshold-filtered panel list — so a low-confidence action/feature the
+// model saw can be typed and accepted in canonical formatting instead of being
+// hand-entered as a custom tag (operator-asked 2026-06-09). The typed query is
+// the only filter; the threshold no longer hides anything here.
 const suggestionHits = computed(() => {
   const q = parsedName.value.toLowerCase()
   if (!q) return []
   const seen = new Set(hits.value.map(h => `${h.kind}:${h.name.toLowerCase()}`))
   const out = []
-  for (const list of Object.values(suggestions.byCategory)) {
+  for (const list of Object.values(suggestions.allByCategory)) {
     for (const s of list || []) {
       const key = `${s.category}:${s.display_name.toLowerCase()}`
       if (!s.display_name.toLowerCase().includes(q)) continue
@@ -204,9 +209,10 @@ const suggestionHits = computed(() => {
       out.push(s)
     }
   }
-  // Best matches first; cap so the dropdown stays scannable.
+  // Best matches first; cap generously so a specific typed query surfaces its
+  // matches even when many predictions exist, while the list stays scrollable.
   out.sort((a, b) => b.score - a.score)
-  return out.slice(0, 6)
+  return out.slice(0, 20)
 })
 
 // One ordered list backing both the rendered dropdown and keyboard nav, so the

frontend/src/stores/suggestions.js

@@ -16,9 +16,14 @@ export const CATEGORY_LABELS = {
 
 export const useSuggestionsStore = defineStore('suggestions', () => {
   const api = useApi()
-  const byCategory = ref({})       // { category: [suggestion, ...] }
+  const byCategory = ref({})       // { category: [suggestion, ...] } — panel (≥ threshold)
+  // The typed tag-input dropdown searches the model's FULL prediction set for
+  // the image (min=0, down to the store floor) so low-confidence actions/
+  // features can be picked in canonical formatting (operator-asked 2026-06-09).
+  const allByCategory = ref({})
   const { loading, error, run } = useAsyncAction({ errorAs: 'message' })
   let currentImageId = null
+  const inflightAll = useInflightToken()
   // Audit 2026-06-02: this store had no inflight guard — a late
   // /suggestions response from a prior image could overwrite
   // byCategory while currentImageId pointed at a new one, and
@@ -42,10 +47,41 @@ export const useSuggestionsStore = defineStore('suggestions', () => {
     })
   }
 
-  function _drop(category, predicate) {
+  // Load the full prediction set for the dropdown (separate inflight guard so a
-    const list = byCategory.value[category]
+  // late response from a prior image can't overwrite the current one's list).
-    if (!list) return
+  async function loadAll(imageId) {
-    byCategory.value[category] = list.filter(s => !predicate(s))
+    inflightAll.cancel()
+    allByCategory.value = {}
+    const t = inflightAll.claim()
+    try {
+      const body = await api.get(`/api/images/${imageId}/suggestions`, {
+        params: { min: 0 },
+      })
+      if (!t.isCurrent()) return
+      allByCategory.value = body.by_category || {}
+    } catch {
+      // Dropdown is best-effort — the panel surfaces load errors.
+    }
+  }
+
+  // A stable identity for a suggestion across the two lists (panel vs dropdown
+  // are separate fetches, so object identity differs): tag id when known, else
+  // the raw display key.
+  function _keyOf(s) {
+    return s.canonical_tag_id != null
+      ? `id:${s.canonical_tag_id}`
+      : `raw:${s.category}:${(s.display_name || '').toLowerCase()}`
+  }
+
+  // Drop an accepted/dismissed suggestion from BOTH the panel list and the
+  // dropdown's full list so it can't reappear in either surface.
+  function _dropEverywhere(suggestion) {
+    const key = _keyOf(suggestion)
+    const cat = suggestion.category
+    for (const map of [byCategory.value, allByCategory.value]) {
+      const list = map[cat]
+      if (list) map[cat] = list.filter(s => _keyOf(s) !== key)
+    }
   }
 
   async function accept(suggestion) {
@@ -70,7 +106,7 @@ export const useSuggestionsStore = defineStore('suggestions', () => {
     // Only drop from THIS image's category list — if the user navigated,
     // the new image has its own suggestions and this drop would corrupt them.
     if (currentImageId === imageId) {
-      _drop(suggestion.category, s => s === suggestion)
+      _dropEverywhere(suggestion)
     }
     toast({
       text: `Tagged: ${suggestion.display_name}`,
@@ -89,7 +125,7 @@ export const useSuggestionsStore = defineStore('suggestions', () => {
       }
     })
     if (currentImageId === imageId) {
-      _drop(suggestion.category, s => s === suggestion)
+      _dropEverywhere(suggestion)
     }
     toast({
       text: `Aliased & tagged: ${suggestion.display_name}`,
@@ -109,12 +145,12 @@ export const useSuggestionsStore = defineStore('suggestions', () => {
       })
     }
     if (currentImageId === imageId) {
-      _drop(suggestion.category, s => s === suggestion)
+      _dropEverywhere(suggestion)
     }
   }
 
   return {
-    byCategory, loading, error,
+    byCategory, allByCategory, loading, error,
-    load, accept, aliasAccept, dismiss
+    load, loadAll, accept, aliasAccept, dismiss
   }
 })

tests/test_ml_suggestions.py

@@ -44,6 +44,33 @@ async def test_threshold_filters_low_confidence_general(db):
     assert "Lowconf" not in names
 
 
+@pytest.mark.asyncio
+async def test_threshold_override_surfaces_low_confidence(db):
+    # The typed-dropdown "show everything the model saw" mode: threshold_override
+    # surfaces stored predictions below the configured threshold (in canonical
+    # formatting) so they can be picked instead of hand-typed (2026-06-09).
+    img = _img(
+        "d" * 64,
+        {
+            "lowconf": {"category": "general", "confidence": 0.30},
+            "sword": {"category": "general", "confidence": 0.97},
+        },
+    )
+    db.add(img)
+    await db.flush()
+    sl = await SuggestionService(db).for_image(img.id, threshold_override=0.0)
+    names = [s.display_name for s in sl.by_category.get("general", [])]
+    assert "Sword" in names
+    assert "Lowconf" in names  # below the configured threshold, surfaced anyway
+
+    # Unsurfaced categories are still excluded even with the override.
+    img2 = _img("e" * 64, {"safe": {"category": "rating", "confidence": 0.99}})
+    db.add(img2)
+    await db.flush()
+    sl2 = await SuggestionService(db).for_image(img2.id, threshold_override=0.0)
+    assert "rating" not in sl2.by_category
+
+
 @pytest.mark.asyncio
 async def test_unsurfaced_category_dropped(db):
     img = _img(