diff --git a/api/__init__.py b/api/__init__.py
index acca117..ece4b75 100644
--- a/api/__init__.py
+++ b/api/__init__.py
@@ -18,7 +18,7 @@ logger = get_logger("api_v1")
 # ---------------------------------------------------------------------------
 limiter = Limiter(
     key_func=get_remote_address,
-    default_limits=["60 per minute"],
+    default_limits=[],           # No global default — limits are applied per-blueprint
     storage_uri="memory://",
 )
 
@@ -43,6 +43,9 @@ def create_api_blueprint():
     from .listenbrainz import register_routes as reg_listenbrainz
     from .cache import register_routes as reg_cache
 
+    # ---- rate-limit only /api/v1 routes (not the whole app) ----
+    limiter.limit("60 per minute")(bp)
+
     reg_library(bp)
     reg_system(bp)
     reg_search(bp)
diff --git a/web_server.py b/web_server.py
index 9c50030..0e24ac3 100644
--- a/web_server.py
+++ b/web_server.py
@@ -20371,7 +20371,7 @@ def update_tidal_discovery_match():
         result['status'] = '✅ Found'
         result['status_class'] = 'found'
         result['spotify_track'] = spotify_track['name']
-        result['spotify_artist'] = ', '.join(spotify_track['artists']) if isinstance(spotify_track['artists'], list) else spotify_track['artists']
+        result['spotify_artist'] = _join_artist_names(spotify_track['artists']) if isinstance(spotify_track['artists'], list) else _extract_artist_name(spotify_track['artists'])
         result['spotify_album'] = spotify_track['album']
         result['spotify_id'] = spotify_track['id']
 
@@ -20812,10 +20812,24 @@ def _run_playlist_discovery_worker(playlists, automation_id=None):
                                      log_line=f'Error: {str(e)}', log_type='error')
 
 
+def _extract_artist_name(artist):
+    """Extract artist name string from either a string or dict ({"name": "..."}) format."""
+    if isinstance(artist, dict):
+        return artist.get('name', '')
+    return artist or ''
+
+def _extract_artist_names(artists):
+    """Extract a list of artist name strings from a list that may contain dicts or strings."""
+    return [_extract_artist_name(a) for a in (artists or [])]
+
+def _join_artist_names(artists):
+    """Join artist names from a list that may contain dicts or strings."""
+    return ', '.join(_extract_artist_names(artists))
+
 def _get_discovery_cache_key(title, artist):
     """Normalize title/artist for discovery cache lookup using matching_engine."""
     norm_title = matching_engine.clean_title(title)
-    norm_artist = matching_engine.clean_artist(artist)
+    norm_artist = matching_engine.clean_artist(_extract_artist_name(artist))
     return (norm_title, norm_artist)
 
 
@@ -20835,6 +20849,11 @@ def _validate_discovery_cache_artist(source_artist, cached_match):
     for cand_artist in cached_artists:
         if not cand_artist:
             continue
+        # Handle both string artists and dict artists ({"name": "..."})
+        if isinstance(cand_artist, dict):
+            cand_artist = cand_artist.get('name', '')
+            if not cand_artist:
+                continue
         cand_normalized = matching_engine.normalize_string(cand_artist)
         if source_artist_cleaned in cand_normalized:
             return True
@@ -21585,7 +21604,7 @@ def update_youtube_discovery_match():
         result['status'] = '✅ Found'
         result['status_class'] = 'found'
         result['spotify_track'] = spotify_track['name']
-        result['spotify_artist'] = ', '.join(spotify_track['artists']) if isinstance(spotify_track['artists'], list) else spotify_track['artists']
+        result['spotify_artist'] = _join_artist_names(spotify_track['artists']) if isinstance(spotify_track['artists'], list) else _extract_artist_name(spotify_track['artists'])
         result['spotify_album'] = spotify_track['album']
         result['spotify_id'] = spotify_track['id']
 
@@ -21669,7 +21688,7 @@ def _run_youtube_discovery_worker(url_hash):
                             'status': '✅ Found',
                             'status_class': 'found',
                             'spotify_track': cached_match.get('name', ''),
-                            'spotify_artist': cached_match.get('artists', [''])[0] if cached_match.get('artists') else '',
+                            'spotify_artist': _extract_artist_name(cached_match.get('artists', [''])[0]) if cached_match.get('artists') else '',
                             'spotify_album': cached_match.get('album', {}).get('name', '') if isinstance(cached_match.get('album'), dict) else cached_match.get('album', ''),
                             'duration': f"{track['duration_ms'] // 60000}:{(track['duration_ms'] % 60000) // 1000:02d}" if track['duration_ms'] else '0:00',
                             'discovery_source': discovery_source,
@@ -21807,7 +21826,7 @@ def _run_youtube_discovery_worker(url_hash):
                     'status': '✅ Found' if matched_track else '❌ Not Found',
                     'status_class': 'found' if matched_track else 'not-found',
                     'spotify_track': matched_track.name if matched_track else '',
-                    'spotify_artist': matched_track.artists[0] if matched_track else '',
+                    'spotify_artist': _extract_artist_name(matched_track.artists[0]) if matched_track else '',
                     'spotify_album': matched_track.album if matched_track else '',
                     'duration': f"{track['duration_ms'] // 60000}:{(track['duration_ms'] % 60000) // 1000:02d}" if track['duration_ms'] else '0:00',
                     'discovery_source': discovery_source,
@@ -21932,7 +21951,7 @@ def _run_listenbrainz_discovery_worker(playlist_mbid):
                             'status': '✅ Found',
                             'status_class': 'found',
                             'spotify_track': cached_match.get('name', ''),
-                            'spotify_artist': cached_match.get('artists', [''])[0] if cached_match.get('artists') else '',
+                            'spotify_artist': _extract_artist_name(cached_match.get('artists', [''])[0]) if cached_match.get('artists') else '',
                             'spotify_album': cached_match.get('album', {}).get('name', '') if isinstance(cached_match.get('album'), dict) else cached_match.get('album', ''),
                             'duration': f"{duration_ms // 60000}:{(duration_ms % 60000) // 1000:02d}" if duration_ms else '0:00',
                             'discovery_source': discovery_source,
@@ -22069,7 +22088,7 @@ def _run_listenbrainz_discovery_worker(playlist_mbid):
                     'status': '✅ Found' if matched_track else '❌ Not Found',
                     'status_class': 'found' if matched_track else 'not-found',
                     'spotify_track': matched_track.name if matched_track else '',
-                    'spotify_artist': matched_track.artists[0] if matched_track else '',
+                    'spotify_artist': _extract_artist_name(matched_track.artists[0]) if matched_track else '',
                     'spotify_album': matched_track.album if matched_track else '',
                     'duration': f"{duration_ms // 60000}:{(duration_ms % 60000) // 1000:02d}" if duration_ms else '0:00',
                     'discovery_source': discovery_source,
@@ -27033,7 +27052,7 @@ def update_listenbrainz_discovery_match():
             result['spotify_track'] = spotify_track.get('name', '') if spotify_track else ''
             # Join all artists (matching YouTube/Tidal/Beatport format)
             artists = spotify_track.get('artists', []) if spotify_track else []
-            result['spotify_artist'] = ', '.join(artists) if isinstance(artists, list) else artists
+            result['spotify_artist'] = _join_artist_names(artists) if isinstance(artists, list) else _extract_artist_name(artists)
             # Album comes as a string from the frontend fix modal
             album = spotify_track.get('album', '') if spotify_track else ''
             result['spotify_album'] = album if isinstance(album, str) else album.get('name', '') if isinstance(album, dict) else ''
@@ -28944,7 +28963,7 @@ def update_beatport_discovery_match():
         result['status'] = '✅ Found'
         result['status_class'] = 'found'
         result['spotify_track'] = spotify_track['name']
-        result['spotify_artist'] = ', '.join(spotify_track['artists']) if isinstance(spotify_track['artists'], list) else spotify_track['artists']
+        result['spotify_artist'] = _join_artist_names(spotify_track['artists']) if isinstance(spotify_track['artists'], list) else _extract_artist_name(spotify_track['artists'])
         result['spotify_album'] = spotify_track['album']
         result['spotify_id'] = spotify_track['id']
 
diff --git a/webui/static/script.js b/webui/static/script.js
index b1c1194..ee61b27 100644
--- a/webui/static/script.js
+++ b/webui/static/script.js
@@ -22292,6 +22292,12 @@ async function startYouTubeDiscovery(urlHash) {
             return;
         }
 
+        // Update frontend phase to match backend
+        const state = listenbrainzPlaylistStates[urlHash] || youtubePlaylistStates[urlHash];
+        if (state) {
+            state.phase = 'discovering';
+        }
+
         // Start polling for progress
         startYouTubeDiscoveryPolling(urlHash);
 
@@ -22570,9 +22576,15 @@ function openYouTubeDiscoveryModal(urlHash) {
 
         // Set initial progress if we have discovery results
         if (state.discoveryResults && state.discoveryResults.length > 0) {
+            // Compute progress from results if discoveryProgress is missing/zero
+            let progress = state.discoveryProgress || 0;
+            const matches = state.spotifyMatches || 0;
+            if (progress === 0 && state.discoveryResults.length > 0 && state.playlist.tracks.length > 0) {
+                progress = Math.min(100, Math.round((state.discoveryResults.length / state.playlist.tracks.length) * 100));
+            }
             const progressData = {
-                progress: state.discoveryProgress || 0,
-                spotify_matches: state.spotifyMatches || 0,
+                progress: progress,
+                spotify_matches: matches || state.discoveryResults.filter(r => r.status_class === 'found').length,
                 spotify_total: state.playlist.tracks.length,
                 results: state.discoveryResults
             };
@@ -42422,7 +42434,9 @@ function renderMirroredCard(p, container) {
     `;
     card.addEventListener('click', () => {
         const st = youtubePlaylistStates[hash];
-        if (st && st.phase && st.phase !== 'fresh') {
+        // Treat as non-fresh if phase is set, or if a poller/discovery modal exists
+        const hasActiveDiscovery = activeYouTubePollers[hash] || document.getElementById(`youtube-discovery-modal-${hash}`);
+        if (st && ((st.phase && st.phase !== 'fresh') || hasActiveDiscovery)) {
             if (st.phase === 'downloading' || st.phase === 'download_complete') {
                 // Open download modal directly (follows Tidal/YouTube card click pattern)
                 const spotifyPlaylistId = st.convertedSpotifyPlaylistId;
@@ -42775,9 +42789,11 @@ async function clearMirroredDiscovery(playlistId, name) {
         const data = await res.json();
         if (data.success) {
             showToast(`Cleared discovery for ${name} (${data.cleared} tracks)`, 'success');
-            // Also clear the discovery state so the card goes back to fresh
+            // Also clear the discovery state and remove stale modal DOM
             const hash = `mirrored_${playlistId}`;
             delete youtubePlaylistStates[hash];
+            const staleModal = document.getElementById(`youtube-discovery-modal-${hash}`);
+            if (staleModal) staleModal.remove();
             loadMirroredPlaylists();
         } else {
             showToast(data.error || 'Failed to clear discovery', 'error');
@@ -43240,9 +43256,10 @@ async function discoverMirroredPlaylist(playlistId) {
 
     // If state already exists (discovery in progress or completed), just reopen the modal
     const existingState = youtubePlaylistStates[tempHash];
-    if (existingState && existingState.phase !== 'fresh') {
+    const hasActiveDiscovery = activeYouTubePollers[tempHash] || document.getElementById(`youtube-discovery-modal-${tempHash}`);
+    if (existingState && (existingState.phase !== 'fresh' || hasActiveDiscovery)) {
         openYouTubeDiscoveryModal(tempHash);
-        // Resume polling if discovery is in progress
+        // Resume polling if discovery is in progress but poller stopped
         if (existingState.phase === 'discovering' && !activeYouTubePollers[tempHash]) {
             startYouTubeDiscoveryPolling(tempHash);
         }
diff --git a/webui/static/style.css b/webui/static/style.css
index 32f34b9..860fd01 100644
--- a/webui/static/style.css
+++ b/webui/static/style.css
@@ -28886,23 +28886,45 @@ body {
     .automations-grid { grid-template-columns: 1fr; }
 }
 
-/* --- New Automation Button --- */
+/* --- New Automation Button (rainbow glow) --- */
+@keyframes btn-rainbow-glow {
+    0%   { box-shadow: 0 4px 18px rgba(255, 0, 100, 0.3), 0 2px 6px rgba(0,0,0,0.2); border-color: rgba(255, 0, 100, 0.35); }
+    20%  { box-shadow: 0 4px 18px rgba(255, 140, 0, 0.3), 0 2px 6px rgba(0,0,0,0.2); border-color: rgba(255, 140, 0, 0.35); }
+    40%  { box-shadow: 0 4px 18px rgba(0, 210, 120, 0.3), 0 2px 6px rgba(0,0,0,0.2); border-color: rgba(0, 210, 120, 0.35); }
+    60%  { box-shadow: 0 4px 18px rgba(0, 140, 255, 0.3), 0 2px 6px rgba(0,0,0,0.2); border-color: rgba(0, 140, 255, 0.35); }
+    80%  { box-shadow: 0 4px 18px rgba(160, 0, 255, 0.3), 0 2px 6px rgba(0,0,0,0.2); border-color: rgba(160, 0, 255, 0.35); }
+    100% { box-shadow: 0 4px 18px rgba(255, 0, 100, 0.3), 0 2px 6px rgba(0,0,0,0.2); border-color: rgba(255, 0, 100, 0.35); }
+}
+@keyframes btn-rainbow-glow-hover {
+    0%   { box-shadow: 0 6px 28px rgba(255, 0, 100, 0.45), 0 3px 10px rgba(0,0,0,0.25); border-color: rgba(255, 0, 100, 0.5); }
+    20%  { box-shadow: 0 6px 28px rgba(255, 140, 0, 0.45), 0 3px 10px rgba(0,0,0,0.25); border-color: rgba(255, 140, 0, 0.5); }
+    40%  { box-shadow: 0 6px 28px rgba(0, 210, 120, 0.45), 0 3px 10px rgba(0,0,0,0.25); border-color: rgba(0, 210, 120, 0.5); }
+    60%  { box-shadow: 0 6px 28px rgba(0, 140, 255, 0.45), 0 3px 10px rgba(0,0,0,0.25); border-color: rgba(0, 140, 255, 0.5); }
+    80%  { box-shadow: 0 6px 28px rgba(160, 0, 255, 0.45), 0 3px 10px rgba(0,0,0,0.25); border-color: rgba(160, 0, 255, 0.5); }
+    100% { box-shadow: 0 6px 28px rgba(255, 0, 100, 0.45), 0 3px 10px rgba(0,0,0,0.25); border-color: rgba(255, 0, 100, 0.5); }
+}
 .auto-new-btn {
-    background: linear-gradient(135deg, rgb(var(--accent-rgb)) 0%, rgb(var(--accent-light-rgb)) 100%);
+    position: relative;
+    background: linear-gradient(135deg, rgba(255,255,255,0.10) 0%, rgba(255,255,255,0.04) 100%);
     color: #fff;
-    border: none;
+    border: 1.5px solid rgba(255,255,255,0.15);
     padding: 10px 22px;
     border-radius: 12px;
     font-size: 14px;
     font-weight: 600;
     cursor: pointer;
-    transition: all 0.3s ease;
+    transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
     letter-spacing: 0.3px;
+    animation: btn-rainbow-glow 6s linear infinite;
+    backdrop-filter: blur(12px);
 }
 .auto-new-btn:hover {
     transform: translateY(-2px);
-    box-shadow: 0 8px 24px rgba(var(--accent-rgb), 0.4);
-    filter: brightness(1.1);
+    background: linear-gradient(135deg, rgba(255,255,255,0.16) 0%, rgba(255,255,255,0.08) 100%);
+    animation: btn-rainbow-glow-hover 4s linear infinite;
+}
+.auto-new-btn:active {
+    transform: translateY(0) scale(0.97);
 }
 
 /* --- Stats Summary Bar --- */