fix: normalize compact quest aliases

Co-authored-by: Codex <codex@openai.com>

README.md

@@ -142,9 +142,10 @@ be a normal directory such as `.cache/advisor-dashboard`. Each refresh writes
 updates `latest.json` through an atomic swap. Quest analysis is cached per project using the metadata-first inference
 prompt hash, taxonomy hash, MiniCPM model id, adapter id/revision, local adapter digest, and generation config.
 
-The app starts an hourly scheduler when `ADVISOR_CACHE_DIR` is configured. Manual and scheduled refreshes both acquire
-`$ADVISOR_CACHE_DIR/refresh.lock`, heartbeat while active, and leave the current validated dashboard in place if a new
-run fails validation.
+The app can start an hourly scheduler when `ADVISOR_CACHE_DIR` is configured. The submitted Space keeps scheduled
+refresh disabled and updates the mounted bucket from local builds, so index construction and quest analysis do not run
+inside the public demo runtime. Manual and scheduled refreshes still acquire `$ADVISOR_CACHE_DIR/refresh.lock`,
+heartbeat while active, and leave the current validated dashboard in place if a new run fails validation.
 
 ## Models And Data
 
@@ -303,6 +304,7 @@ ADVISOR_QUEST_ANALYZER_BACKEND=minicpm-transformers
 ADVISOR_QUEST_ADAPTER_ID=build-small-hackathon/hackathon-advisor-quest-minicpm5-lora
 ADVISOR_QUEST_ANALYSIS_BATCH_SIZE=8
 ADVISOR_CACHE_DIR=/data/advisor-cache
+ADVISOR_DISABLE_SCHEDULED_REFRESH=1
 ADVISOR_REFRESH_COMPUTE=cpu
 ADVISOR_SCHEDULED_REFRESH_COMPUTE=cpu
 ADVISOR_REFRESH_INTERVAL_SECONDS=3600

hackathon_advisor/quest_taxonomy.py

@@ -172,6 +172,12 @@ _QUEST_ALIASES.update(
         _quest_key("OpenAI Codex"): "Codex",
         _quest_key("OpenAI"): "Codex",
         _quest_key("Codex"): "Codex",
+        _quest_key("Offbrand"): "Off-Brand",
+        _quest_key("Offgrid"): "Off the Grid",
+        _quest_key("Llama Champion badge"): "Llama Champion",
+        _quest_key("Modal first"): "Modal",
+        _quest_key("Nemotron 3 Nano 4B"): "Nemotron",
+        _quest_key("Nemotron Mini 4B"): "Nemotron",
         _quest_key("Small model <=4B"): "Tiny Titan",
         _quest_key("Small model under 4B"): "Tiny Titan",
         _quest_key("Shareable output"): "Sharing is Caring",

tests/test_dashboard.py

@@ -308,6 +308,54 @@ def test_quest_analysis_validation_accepts_best_use_of_known_labels() -> None:
     assert validated.matches_by_project[projects[0].id][0]["quest"] == "Modal"
 
 
+def test_quest_analysis_validation_accepts_common_compact_aliases() -> None:
+    projects = fake_projects(1)
+    raw = {
+        "projects": [
+            {
+                "project_id": projects[0].id,
+                "matches": [
+                    {
+                        "quest": "Offbrand",
+                        "confidence": 0.84,
+                        "evidence": "custom frontend",
+                        "source": "readme",
+                    },
+                    {
+                        "quest": "Offgrid",
+                        "confidence": 0.82,
+                        "evidence": "loads weights locally",
+                        "source": "app_file",
+                    },
+                    {
+                        "quest": "Llama Champion badge",
+                        "confidence": 0.8,
+                        "evidence": "llama-cpp-python",
+                        "source": "app_file",
+                    },
+                    {
+                        "quest": "Modal-first",
+                        "confidence": 0.79,
+                        "evidence": "Modal serverless GPUs",
+                        "source": "readme",
+                    },
+                    {
+                        "quest": "Nemotron-3 Nano 4B",
+                        "confidence": 0.78,
+                        "evidence": "Nemotron 3 Nano 4B",
+                        "source": "readme",
+                    },
+                ],
+            }
+        ]
+    }
+
+    validated = validate_quest_analysis_payload(raw, projects, source="fake")
+
+    quests = [match["quest"] for match in validated.matches_by_project[projects[0].id]]
+    assert quests == ["Off-Brand", "Off the Grid", "Llama Champion", "Modal", "Nemotron"]
+
+
 def test_quest_analysis_validation_rejects_unknown_composite_quest_labels() -> None:
     projects = fake_projects(1)
     raw = {