feat: 8 AI providers, rich text editor, agent notifications, UI contrast & font settings

- Add DeepSeek, OpenRouter, Mistral, Z.AI, LM Studio as AI providers
  with editable model names via Combobox in admin settings
- Fix OpenRouter broken by normalizeProvider bug in config.ts
- Convert agent-created notes from Markdown to HTML (TipTap rich text)
- Add Notification model + in-app notifications for agent results
- Agent notification click opens the created note directly
- Add note count display on notebook and inbox headers
- Fix checklist toggle in card view (persist state via localCheckItems)
- Add checklist creation option in tabs/list view (dropdown on + button)
- Fix image description ENOENT error with HTTP fallback
- Improve UI contrast across all themes (input, border, checkbox visibility)
- Add font family setting (Inter vs System Default) in Appearance settings
- Fix CSS font-sans variable conflict (removed dead Geist references)
- Update README with new features and 8 providers

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

memento-note/app/api/admin/models/route.ts

@@ -10,13 +10,13 @@ async function requireAdmin() {
 /**
  * GET /api/admin/models?type=ollama&url=<base_url>
  * GET /api/admin/models?type=custom&url=<base_url>&key=<api_key>&kind=tags|embeddings
+ * GET /api/admin/models?type=deepseek&key=<api_key>&kind=tags|embeddings
+ * GET /api/admin/models?type=openrouter&key=<api_key>&kind=tags|embeddings
+ * GET /api/admin/models?type=mistral&key=<api_key>&kind=tags|embeddings
+ * GET /api/admin/models?type=zai&key=<api_key>&kind=tags|embeddings
+ * GET /api/admin/models?type=lmstudio&url=<base_url>
  *
- * Route API (not a Server Action) for fetching AI model lists from Ollama or
- * OpenAI-compatible providers. Using a Route Handler instead of a Server Action
- * is the correct architecture for client-side GET requests: Server Actions are
- * for data mutations, and calling them from useEffect pushes items into the
- * App Router's internal action queue, which is drained during render (inside
- * AppRouter's useMemo), triggering React Error #310 when multiple calls stack up.
+ * Route API for fetching AI model lists from providers.
  */
 export async function GET(request: NextRequest) {
   if (!(await requireAdmin())) {
@@ -29,14 +29,22 @@ export async function GET(request: NextRequest) {
   const apiKey = searchParams.get('key') ?? undefined
   const kind = searchParams.get('kind') ?? 'tags'
 
-  if (!rawUrl) {
-    return NextResponse.json({ success: false, models: [], error: 'url parameter is required' })
+  // Provider-specific base URLs (used when url param is empty)
+  const PROVIDER_URLS: Record<string, string> = {
+    deepseek: 'https://api.deepseek.com/v1',
+    openrouter: 'https://openrouter.ai/api/v1',
+    mistral: 'https://api.mistral.ai/v1',
+    zai: 'https://api.zukijourney.com/v1',
+    lmstudio: 'http://localhost:1234/v1',
   }
 
-  const baseUrl = rawUrl.replace(/\/$/, '').replace(/\/v1$/, '')
-
   try {
+    // Ollama: uses native /api/tags endpoint
     if (type === 'ollama') {
+      if (!rawUrl) {
+        return NextResponse.json({ success: false, models: [], error: 'url parameter is required for Ollama' })
+      }
+      const baseUrl = rawUrl.replace(/\/$/, '').replace(/\/api$/, '')
       const res = await fetch(`${baseUrl}/api/tags`, {
         headers: { 'Content-Type': 'application/json' },
         signal: AbortSignal.timeout(5000),
@@ -49,58 +57,69 @@ export async function GET(request: NextRequest) {
       return NextResponse.json({ success: true, models })
     }
 
-    if (type === 'custom') {
-      const headers: Record<string, string> = { 'Content-Type': 'application/json' }
-      if (apiKey) headers['Authorization'] = `Bearer ${apiKey}`
+    // All other providers: use OpenAI-compatible /v1/models endpoint
+    const baseUrl = rawUrl
+      ? rawUrl.replace(/\/$/, '')
+      : (PROVIDER_URLS[type || ''] || '')
 
-      if (kind === 'embeddings') {
-        // Try provider-specific embeddings endpoint first (e.g. OpenRouter)
-        try {
-          const embRes = await fetch(`${baseUrl}/v1/embeddings/models`, {
-            headers,
-            signal: AbortSignal.timeout(8000),
-          })
-          if (embRes.ok) {
-            const embData = await embRes.json()
-            const embModels: string[] = (embData.data ?? [])
-              .map((m: { id: string }) => m.id)
-              .filter(Boolean)
-              .sort()
-            if (embModels.length > 0) {
-              return NextResponse.json({ success: true, models: embModels })
-            }
-          }
-        } catch {
-          // Fall through to /v1/models with keyword filter
-        }
-      }
-
-      const res = await fetch(`${baseUrl}/v1/models`, {
-        headers,
-        signal: AbortSignal.timeout(8000),
-      })
-      if (!res.ok) {
-        return NextResponse.json({ success: false, models: [], error: `Provider ${res.status}` })
-      }
-
-      const data = await res.json()
-      let models: string[] = (data.data ?? [])
-        .map((m: { id: string }) => m.id)
-        .filter(Boolean)
-        .sort()
-
-      if (kind === 'embeddings') {
-        const keywords = ['embed', 'embedding', 'ada', 'e5', 'bge', 'gte', 'minilm']
-        const filtered = models.filter((id) =>
-          keywords.some((kw) => id.toLowerCase().includes(kw))
-        )
-        if (filtered.length > 0) models = filtered
-      }
-
-      return NextResponse.json({ success: true, models })
+    if (!baseUrl) {
+      return NextResponse.json({ success: false, models: [], error: 'url parameter is required' })
     }
 
-    return NextResponse.json({ success: false, models: [], error: `Unknown type: ${type}` })
+    const headers: Record<string, string> = { 'Content-Type': 'application/json' }
+    if (apiKey) headers['Authorization'] = `Bearer ${apiKey}`
+
+    // For OpenRouter, add required headers
+    if (type === 'openrouter') {
+      headers['HTTP-Referer'] = 'https://localhost:3000'
+      headers['X-Title'] = 'Memento AI'
+    }
+
+    // Try provider-specific embeddings endpoint first for embeddings kind
+    if (kind === 'embeddings') {
+      try {
+        const embRes = await fetch(`${baseUrl}/embeddings/models`, {
+          headers,
+          signal: AbortSignal.timeout(8000),
+        })
+        if (embRes.ok) {
+          const embData = await embRes.json()
+          const embModels: string[] = (embData.data ?? [])
+            .map((m: { id: string }) => m.id)
+            .filter(Boolean)
+            .sort()
+          if (embModels.length > 0) {
+            return NextResponse.json({ success: true, models: embModels })
+          }
+        }
+      } catch {
+        // Fall through to /v1/models with keyword filter
+      }
+    }
+
+    const res = await fetch(`${baseUrl}/models`, {
+      headers,
+      signal: AbortSignal.timeout(8000),
+    })
+    if (!res.ok) {
+      return NextResponse.json({ success: false, models: [], error: `Provider ${res.status}` })
+    }
+
+    const data = await res.json()
+    let models: string[] = (data.data ?? [])
+      .map((m: { id: string }) => m.id)
+      .filter(Boolean)
+      .sort()
+
+    if (kind === 'embeddings') {
+      const keywords = ['embed', 'embedding', 'ada', 'e5', 'bge', 'gte', 'minilm']
+      const filtered = models.filter((id) =>
+        keywords.some((kw) => id.toLowerCase().includes(kw))
+      )
+      if (filtered.length > 0) models = filtered
+    }
+
+    return NextResponse.json({ success: true, models })
   } catch (err: any) {
     return NextResponse.json({ success: false, models: [], error: err.message ?? 'Request failed' })
   }