feat: Implement multi-provider AI support

Backend:
- Add per-provider API key fields to AIConfig (AnthropicAPIKey, OpenAIAPIKey, DeepSeekAPIKey, OllamaBaseURL, OpenAIBaseURL)
- Add NewForProvider() and NewForModel() factory functions for multi-provider instantiation
- Update ListModels() to aggregate models from all configured providers with provider:model format
- Update Execute/ExecuteStream to dynamically create provider based on selected model
- Update TestConnection to use multi-provider aware provider creation
- Add helper functions: HasProvider(), GetConfiguredProviders(), GetAPIKeyForProvider(), GetBaseURLForProvider(), ParseModelString(), FormatModelString()

Frontend:
- Remove legacy single-provider UI (provider grid, single API key input, single base URL)
- Add accordion-style UI for configuring all providers independently
- Add model grouping by provider in selectors using optgroup
- Update AIChat model dropdown with grouped provider sections
- Add helper functions for parsing provider from model ID and grouping models

API:
- Add multi-provider fields to AISettingsResponse and AISettingsUpdateRequest
- Add /api/ai/models endpoint for dynamic model listing
- Update settings handlers for per-provider credential management

frontend-modern/src/api/ai.ts

@@ -32,6 +32,11 @@ export class AIAPI {
     }) as Promise<AITestResult>;
   }
 
+  // Get available models from the AI provider
+  static async getModels(): Promise<{ models: { id: string; name: string; description?: string }[]; error?: string }> {
+    return apiFetchJSON(`${this.baseUrl}/ai/models`) as Promise<{ models: { id: string; name: string; description?: string }[]; error?: string }>;
+  }
+
   // Start OAuth flow for Claude Pro/Max subscription
   // Returns the authorization URL to redirect the user to
   static async startOAuth(): Promise<{ auth_url: string; state: string }> {

frontend-modern/src/api/patrol.ts

@@ -61,7 +61,7 @@ export interface PatrolRunRecord {
     started_at: string;
     completed_at: string;
     duration_ms: number;
-    type: 'quick' | 'deep';
+    type: string; // Always 'patrol' now (kept for backwards compat with old records)
     resources_checked: number;
     // Breakdown by resource type
     nodes_checked: number;

frontend-modern/src/components/AI/AIChat.tsx

@@ -1,4 +1,4 @@
-import { Component, Show, createSignal, For, createEffect, createMemo } from 'solid-js';
+import { Component, Show, createSignal, For, createEffect, createMemo, onMount } from 'solid-js';
 import { marked } from 'marked';
 import { AIAPI } from '@/api/ai';
 import { notificationStore } from '@/stores/notifications';
@@ -13,8 +13,50 @@ import type {
   AIStreamToolEndData,
   AIStreamCompleteData,
   AIStreamApprovalNeededData,
+  ModelInfo,
 } from '@/types/ai';
 
+// Provider display names for grouped model selection
+const PROVIDER_DISPLAY_NAMES: Record<string, string> = {
+  anthropic: 'Anthropic (Claude)',
+  openai: 'OpenAI (GPT)',
+  deepseek: 'DeepSeek',
+  ollama: 'Ollama (Local)',
+};
+
+// Parse provider from model ID (format: "provider:model-name")
+function getProviderFromModelId(modelId: string): string {
+  const colonIndex = modelId.indexOf(':');
+  if (colonIndex > 0) {
+    return modelId.substring(0, colonIndex);
+  }
+  // Default detection for models without prefix
+  if (modelId.includes('claude') || modelId.includes('opus') || modelId.includes('sonnet') || modelId.includes('haiku')) {
+    return 'anthropic';
+  }
+  if (modelId.includes('gpt') || modelId.includes('o1') || modelId.includes('o3')) {
+    return 'openai';
+  }
+  if (modelId.includes('deepseek')) {
+    return 'deepseek';
+  }
+  return 'ollama';
+}
+
+// Group models by provider for grouped rendering
+function groupModelsByProvider(models: ModelInfo[]): Map<string, ModelInfo[]> {
+  const grouped = new Map<string, ModelInfo[]>();
+
+  for (const model of models) {
+    const provider = getProviderFromModelId(model.id);
+    const existing = grouped.get(provider) || [];
+    existing.push(model);
+    grouped.set(provider, existing);
+  }
+
+  return grouped;
+}
+
 // Configure marked for safe rendering
 marked.setOptions({
   breaks: true, // Convert \n to <br>
@@ -116,6 +158,7 @@ export const AIChat: Component<AIChatProps> = (props) => {
   const targetId = () => context().targetId;
   const contextData = () => context().context;
   const initialPrompt = () => context().initialPrompt;
+  const findingId = () => context().findingId; // For resolving patrol findings
 
   // Access WebSocket state for listing available resources
   const wsContext = useWebSocket();
@@ -247,10 +290,32 @@ export const AIChat: Component<AIChatProps> = (props) => {
   const [input, setInput] = createSignal('');
   const [isLoading, setIsLoading] = createSignal(false);
   const [queuedMessage, setQueuedMessage] = createSignal<string | null>(null);
+
+  // Model selection
+  const [availableModels, setAvailableModels] = createSignal<ModelInfo[]>([]);
+  const [selectedModel, setSelectedModel] = createSignal<string>(''); // Empty = use default
+  const [showModelSelector, setShowModelSelector] = createSignal(false);
+
   let messagesEndRef: HTMLDivElement | undefined;
   let inputRef: HTMLTextAreaElement | undefined;
   let abortControllerRef: AbortController | null = null;
 
+  // Fetch available models on mount using the dynamic API
+  onMount(async () => {
+    try {
+      const result = await AIAPI.getModels();
+      if (result.models && result.models.length > 0) {
+        setAvailableModels(result.models.map(m => ({
+          id: m.id,
+          name: m.name || m.id,
+          description: m.description,
+        })));
+      }
+    } catch (e) {
+      // Silently fail - models will just not be selectable
+    }
+  });
+
   // Wrapper to sync messages to global store
   const setMessages = (updater: Message[] | ((prev: Message[]) => Message[])) => {
     setMessagesLocal((prev) => {
@@ -425,6 +490,8 @@ export const AIChat: Component<AIChatProps> = (props) => {
           target_id: targetId(),
           context: contextData(),
           history: history.length > 0 ? history : undefined,
+          finding_id: findingId(), // Pass finding ID so AI can resolve it when fixed
+          model: selectedModel() || undefined, // Use selected model or default
         },
         (event: AIStreamEvent) => {
           lastEventTime = Date.now(); // Update last event time
@@ -871,6 +938,58 @@ export const AIChat: Component<AIChatProps> = (props) => {
             </div>
           </div>
           <div class="flex items-center gap-2">
+            {/* Model selector dropdown */}
+            <Show when={availableModels().length > 0}>
+              <div class="relative">
+                <button
+                  onClick={() => setShowModelSelector(!showModelSelector())}
+                  class="flex items-center gap-1 px-2 py-1 text-xs text-gray-500 dark:text-gray-400 hover:text-gray-700 dark:hover:text-gray-200 rounded border border-gray-200 dark:border-gray-700 hover:border-gray-300 dark:hover:border-gray-600 bg-white dark:bg-gray-800"
+                  title="Select AI model"
+                >
+                  <svg class="w-3 h-3" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                    <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9.75 17L9 20l-1 1h8l-1-1-.75-3M3 13h18M5 17h14a2 2 0 002-2V5a2 2 0 00-2-2H5a2 2 0 00-2 2v10a2 2 0 002 2z" />
+                  </svg>
+                  <span class="max-w-[80px] truncate">
+                    {selectedModel() || 'Default'}
+                  </span>
+                  <svg class="w-3 h-3" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                    <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 9l-7 7-7-7" />
+                  </svg>
+                </button>
+                <Show when={showModelSelector()}>
+                  <div class="absolute right-0 top-full mt-1 w-64 bg-white dark:bg-gray-800 rounded-lg shadow-lg border border-gray-200 dark:border-gray-700 z-50 py-1">
+                    <button
+                      onClick={() => { setSelectedModel(''); setShowModelSelector(false); }}
+                      class={`w-full px-3 py-2 text-left text-sm hover:bg-gray-100 dark:hover:bg-gray-700 ${!selectedModel() ? 'bg-blue-50 dark:bg-blue-900/30' : ''}`}
+                    >
+                      <div class="font-medium text-gray-900 dark:text-gray-100">Default</div>
+                      <div class="text-xs text-gray-500 dark:text-gray-400">Use configured default model</div>
+                    </button>
+                    <For each={Array.from(groupModelsByProvider(availableModels()).entries())}>
+                      {([provider, models]) => (
+                        <>
+                          <div class="px-3 py-1.5 text-xs font-semibold text-gray-500 dark:text-gray-400 bg-gray-50 dark:bg-gray-700/50 sticky top-0">
+                            {PROVIDER_DISPLAY_NAMES[provider] || provider}
+                          </div>
+                          <For each={models}>
+                            {(model) => (
+                              <button
+                                onClick={() => { setSelectedModel(model.id); setShowModelSelector(false); }}
+                                class={`w-full px-3 py-2 text-left text-sm hover:bg-gray-100 dark:hover:bg-gray-700 ${selectedModel() === model.id ? 'bg-blue-50 dark:bg-blue-900/30' : ''}`}
+                              >
+                                <div class="font-medium text-gray-900 dark:text-gray-100">
+                                  {model.name || model.id.split(':').pop()}
+                                </div>
+                              </button>
+                            )}
+                          </For>
+                        </>
+                      )}
+                    </For>
+                  </div>
+                </Show>
+              </div>
+            </Show>
             <button
               onClick={clearChat}
               class="p-2 text-gray-400 hover:text-gray-600 dark:hover:text-gray-300 rounded-lg hover:bg-gray-100 dark:hover:bg-gray-800"

frontend-modern/src/components/Settings/AISettings.tsx

@@ -8,36 +8,84 @@ import { notificationStore } from '@/stores/notifications';
 import { logger } from '@/utils/logger';
 import { AIAPI } from '@/api/ai';
 import type { AISettings as AISettingsType, AIProvider, AuthMethod } from '@/types/ai';
-import { PROVIDER_NAMES, PROVIDER_DESCRIPTIONS, DEFAULT_MODELS } from '@/types/ai';
+import { DEFAULT_MODELS } from '@/types/ai';
 
-const PROVIDERS: AIProvider[] = ['anthropic', 'openai', 'ollama', 'deepseek'];
+// Providers are now configured via accordion sections, not a single-provider selector
+
+// Provider display names for optgroup labels
+const PROVIDER_DISPLAY_NAMES: Record<string, string> = {
+  anthropic: 'Anthropic (Claude)',
+  openai: 'OpenAI (GPT)',
+  deepseek: 'DeepSeek',
+  ollama: 'Ollama (Local)',
+};
+
+// Parse provider from model ID (format: "provider:model-name")
+function getProviderFromModelId(modelId: string): string {
+  const colonIndex = modelId.indexOf(':');
+  if (colonIndex > 0) {
+    return modelId.substring(0, colonIndex);
+  }
+  // Default detection for models without prefix
+  if (modelId.includes('claude') || modelId.includes('opus') || modelId.includes('sonnet') || modelId.includes('haiku')) {
+    return 'anthropic';
+  }
+  if (modelId.includes('gpt') || modelId.includes('o1') || modelId.includes('o3')) {
+    return 'openai';
+  }
+  if (modelId.includes('deepseek')) {
+    return 'deepseek';
+  }
+  return 'ollama';
+}
+
+// Group models by provider for optgroup rendering
+function groupModelsByProvider(models: { id: string; name: string; description?: string }[]): Map<string, { id: string; name: string; description?: string }[]> {
+  const grouped = new Map<string, { id: string; name: string; description?: string }[]>();
+
+  for (const model of models) {
+    const provider = getProviderFromModelId(model.id);
+    const existing = grouped.get(provider) || [];
+    existing.push(model);
+    grouped.set(provider, existing);
+  }
+
+  return grouped;
+}
 
 export const AISettings: Component = () => {
   const [settings, setSettings] = createSignal<AISettingsType | null>(null);
   const [loading, setLoading] = createSignal(false);
   const [saving, setSaving] = createSignal(false);
   const [testing, setTesting] = createSignal(false);
-  const [startingOAuth, setStartingOAuth] = createSignal(false);
-  const [disconnectingOAuth, setDisconnectingOAuth] = createSignal(false);
-  const [exchangingCode, setExchangingCode] = createSignal(false);
 
-  // OAuth flow state
+  // Dynamic model list from provider API
-  const [oauthAuthUrl, setOAuthAuthUrl] = createSignal<string | null>(null);
+  const [availableModels, setAvailableModels] = createSignal<{ id: string; name: string; description?: string }[]>([]);
-  const [oauthState, setOAuthState] = createSignal<string | null>(null);
+  const [modelsLoading, setModelsLoading] = createSignal(false);
-  const [oauthCode, setOAuthCode] = createSignal('');
+
+  // Accordion state for provider configuration sections
+  const [expandedProviders, setExpandedProviders] = createSignal<Set<AIProvider>>(new Set(['anthropic']));
 
   const [form, setForm] = createStore({
     enabled: false,
-    provider: 'anthropic' as AIProvider,
+    provider: 'anthropic' as AIProvider, // Legacy - kept for compatibility
-    apiKey: '',
+    apiKey: '', // Legacy - kept for compatibility
     model: '',
-    baseUrl: '',
+    chatModel: '', // Empty means use default model
+    patrolModel: '', // Empty means use default model
+    baseUrl: '', // Legacy - kept for compatibility
     clearApiKey: false,
     autonomousMode: false,
     authMethod: 'api_key' as AuthMethod,
     patrolSchedulePreset: '6hr',
     alertTriggeredAnalysis: true,
     patrolAutoFix: false,
+    // Multi-provider credentials
+    anthropicApiKey: '',
+    openaiApiKey: '',
+    deepseekApiKey: '',
+    ollamaBaseUrl: 'http://localhost:11434',
+    openaiBaseUrl: '',
   });
 
   const resetForm = (data: AISettingsType | null) => {
@@ -47,6 +95,8 @@ export const AISettings: Component = () => {
         provider: 'anthropic',
         apiKey: '',
         model: DEFAULT_MODELS.anthropic,
+        chatModel: '',
+        patrolModel: '',
         baseUrl: '',
         clearApiKey: false,
         autonomousMode: false,
@@ -54,6 +104,12 @@ export const AISettings: Component = () => {
         patrolSchedulePreset: '6hr',
         alertTriggeredAnalysis: true,
         patrolAutoFix: false,
+        // Multi-provider - empty by default
+        anthropicApiKey: '',
+        openaiApiKey: '',
+        deepseekApiKey: '',
+        ollamaBaseUrl: 'http://localhost:11434',
+        openaiBaseUrl: '',
       });
       return;
     }
@@ -63,6 +119,8 @@ export const AISettings: Component = () => {
       provider: data.provider,
       apiKey: '',
       model: data.model || DEFAULT_MODELS[data.provider],
+      chatModel: data.chat_model || '',
+      patrolModel: data.patrol_model || '',
       baseUrl: data.base_url || '',
       clearApiKey: false,
       autonomousMode: data.autonomous_mode || false,
@@ -70,7 +128,39 @@ export const AISettings: Component = () => {
       patrolSchedulePreset: data.patrol_schedule_preset || '6hr',
       alertTriggeredAnalysis: data.alert_triggered_analysis !== false, // default to true
       patrolAutoFix: data.patrol_auto_fix || false, // default to false (observe only)
+      // Multi-provider - never load actual keys from server (security), just track if configured
+      anthropicApiKey: '', // Always empty - we only show if configured
+      openaiApiKey: '',
+      deepseekApiKey: '',
+      ollamaBaseUrl: data.ollama_base_url || 'http://localhost:11434',
+      openaiBaseUrl: data.openai_base_url || '',
     });
+
+    // Auto-expand providers that are configured
+    const configured = new Set<AIProvider>();
+    if (data.anthropic_configured) configured.add('anthropic');
+    if (data.openai_configured) configured.add('openai');
+    if (data.deepseek_configured) configured.add('deepseek');
+    if (data.ollama_configured) configured.add('ollama');
+    // Default to anthropic if nothing is configured
+    if (configured.size === 0) configured.add('anthropic');
+    setExpandedProviders(configured);
+  };
+
+  // Load available models from the provider's API
+  const loadModels = async () => {
+    setModelsLoading(true);
+    try {
+      const result = await AIAPI.getModels();
+      if (result.models && result.models.length > 0) {
+        setAvailableModels(result.models);
+      }
+    } catch (e) {
+      // Silently fail - user can still type model names manually
+      logger.debug('[AISettings] Failed to load models from API:', e);
+    } finally {
+      setModelsLoading(false);
+    }
   };
 
   const loadSettings = async () => {
@@ -79,6 +169,10 @@ export const AISettings: Component = () => {
       const data = await AIAPI.getSettings();
       setSettings(data);
       resetForm(data);
+      // Load available models after settings (needs API key to be configured)
+      if (data.configured) {
+        loadModels();
+      }
     } catch (error) {
       logger.error('[AISettings] Failed to load settings:', error);
       notificationStore.error('Failed to load AI settings');
@@ -116,14 +210,8 @@ export const AISettings: Component = () => {
     }
   });
 
-  const handleProviderChange = (provider: AIProvider) => {
+  // Note: handleProviderChange is no longer used as we now use multi-provider accordions
-    setForm('provider', provider);
+  // The provider is implicitly determined by the selected model (e.g., "anthropic:claude-opus")
-    // Update model to default for new provider if current model doesn't look like it belongs
-    const currentModel = form.model;
-    if (!currentModel || currentModel === DEFAULT_MODELS[settings()?.provider || 'anthropic']) {
-      setForm('model', DEFAULT_MODELS[provider]);
-    }
-  };
 
   const handleSave = async (event?: Event) => {
     event?.preventDefault();
@@ -170,6 +258,33 @@ export const AISettings: Component = () => {
         payload.patrol_auto_fix = form.patrolAutoFix;
       }
 
+      // Include model overrides if changed
+      if (form.chatModel !== (settings()?.chat_model || '')) {
+        payload.chat_model = form.chatModel;
+      }
+
+      if (form.patrolModel !== (settings()?.patrol_model || '')) {
+        payload.patrol_model = form.patrolModel;
+      }
+
+      // Include multi-provider credentials if set (non-empty)
+      if (form.anthropicApiKey.trim()) {
+        payload.anthropic_api_key = form.anthropicApiKey.trim();
+      }
+      if (form.openaiApiKey.trim()) {
+        payload.openai_api_key = form.openaiApiKey.trim();
+      }
+      if (form.deepseekApiKey.trim()) {
+        payload.deepseek_api_key = form.deepseekApiKey.trim();
+      }
+      // Always include Ollama URL changes
+      if (form.ollamaBaseUrl !== (settings()?.ollama_base_url || 'http://localhost:11434')) {
+        payload.ollama_base_url = form.ollamaBaseUrl.trim();
+      }
+      if (form.openaiBaseUrl !== (settings()?.openai_base_url || '')) {
+        payload.openai_base_url = form.openaiBaseUrl.trim();
+      }
+
       const updated = await AIAPI.updateSettings(payload);
       setSettings(updated);
       resetForm(updated);
@@ -201,82 +316,10 @@ export const AISettings: Component = () => {
     }
   };
 
-  const handleStartOAuth = async () => {
+  // OAuth handlers removed - OAuth is currently unavailable from Anthropic for third-party apps
-    setStartingOAuth(true);
+  // When OAuth becomes available, handlers can be added back to the Anthropic accordion section
-    try {
-      const result = await AIAPI.startOAuth();
-      // Store the auth URL and state for the user to visit manually
-      setOAuthAuthUrl(result.auth_url);
-      setOAuthState(result.state);
-      setOAuthCode('');
-      notificationStore.info('Click the link below to sign in, then paste the code back here', 5000);
-    } catch (error) {
-      logger.error('[AISettings] OAuth start failed:', error);
-      const message = error instanceof Error ? error.message : 'Failed to start OAuth flow';
-      notificationStore.error(message);
-    } finally {
-      setStartingOAuth(false);
-    }
-  };
 
-  const handleExchangeCode = async () => {
+  // Legacy helper functions removed - multi-provider accordions handle all provider-specific UI
-    const code = oauthCode().trim();
-    const state = oauthState();
-
-    if (!code || !state) {
-      notificationStore.error('Please enter the authorization code');
-      return;
-    }
-
-    setExchangingCode(true);
-    try {
-      await AIAPI.exchangeOAuthCode(code, state);
-      notificationStore.success('Successfully connected to Claude with your subscription!');
-      // Clear OAuth flow state
-      setOAuthAuthUrl(null);
-      setOAuthState(null);
-      setOAuthCode('');
-      // Reload settings
-      await loadSettings();
-    } catch (error) {
-      logger.error('[AISettings] OAuth code exchange failed:', error);
-      const message = error instanceof Error ? error.message : 'Failed to exchange authorization code';
-      notificationStore.error(message);
-    } finally {
-      setExchangingCode(false);
-    }
-  };
-
-  const handleCancelOAuth = () => {
-    setOAuthAuthUrl(null);
-    setOAuthState(null);
-    setOAuthCode('');
-  };
-
-  const handleDisconnectOAuth = async () => {
-    if (!confirm('Are you sure you want to disconnect your Claude subscription? You will need to provide an API key to continue using AI features.')) {
-      return;
-    }
-
-    setDisconnectingOAuth(true);
-    try {
-      await AIAPI.disconnectOAuth();
-      notificationStore.success('Claude subscription disconnected');
-      // Reload settings
-      await loadSettings();
-    } catch (error) {
-      logger.error('[AISettings] OAuth disconnect failed:', error);
-      const message = error instanceof Error ? error.message : 'Failed to disconnect OAuth';
-      notificationStore.error(message);
-    } finally {
-      setDisconnectingOAuth(false);
-    }
-  };
-
-  const needsApiKey = () => form.provider !== 'ollama' && (form.provider !== 'anthropic' || form.authMethod !== 'oauth');
-  const showBaseUrl = () => form.provider === 'ollama' || form.provider === 'openai' || form.provider === 'deepseek';
-  const isAnthropicWithOAuth = () => form.provider === 'anthropic' && form.authMethod === 'oauth';
-  const showAuthMethodSelector = () => form.provider === 'anthropic';
 
   return (
     <Card
@@ -342,289 +385,308 @@ export const AISettings: Component = () => {
 
         <Show when={!loading()}>
           <div class="space-y-4">
-            {/* Provider Selection */}
+            {/* Model Selection */}
             <div class={formField}>
-              <label class={labelClass()}>AI Provider</label>
+              <label class={labelClass()}>
-              <div class="grid grid-cols-3 gap-3">
+                Default Model
-                <For each={PROVIDERS}>
+                {modelsLoading() && <span class="ml-2 text-xs text-gray-500">(loading models...)</span>}
-                  {(provider) => (
+              </label>
-                    <button
+              <Show when={availableModels().length > 0} fallback={
-                      type="button"
-                      class={`p-3 rounded-lg border-2 text-left transition-all ${form.provider === provider
-                        ? 'border-purple-500 bg-purple-50 dark:bg-purple-900/30'
-                        : 'border-gray-200 dark:border-gray-700 hover:border-gray-300 dark:hover:border-gray-600'
-                        }`}
-                      onClick={() => handleProviderChange(provider)}
-                      disabled={saving()}
-                    >
-                      <div class="font-medium text-sm text-gray-900 dark:text-gray-100">
-                        {PROVIDER_NAMES[provider]}
-                      </div>
-                      <div class="text-xs text-gray-500 dark:text-gray-400 mt-0.5">
-                        {PROVIDER_DESCRIPTIONS[provider]}
-                      </div>
-                    </button>
-                  )}
-                </For>
-              </div>
-            </div>
-
-            {/* Authentication Method - only for Anthropic */}
-            <Show when={showAuthMethodSelector()}>
-              <div class={formField}>
-                <label class={labelClass()}>Authentication Method</label>
-                <div class="grid grid-cols-2 gap-3">
-                  <button
-                    type="button"
-                    class={`p-3 rounded-lg border-2 text-left transition-all ${form.authMethod === 'api_key'
-                      ? 'border-purple-500 bg-purple-50 dark:bg-purple-900/30'
-                      : 'border-gray-200 dark:border-gray-700 hover:border-gray-300 dark:hover:border-gray-600'
-                      }`}
-                    onClick={() => setForm('authMethod', 'api_key')}
-                    disabled={saving()}
-                  >
-                    <div class="font-medium text-sm text-gray-900 dark:text-gray-100">
-                      API Key
-                    </div>
-                    <div class="text-xs text-gray-500 dark:text-gray-400 mt-0.5">
-                      Pay per token usage
-                    </div>
-                  </button>
-                  <button
-                    type="button"
-                    class={`p-3 rounded-lg border-2 text-left transition-all opacity-60 cursor-not-allowed ${form.authMethod === 'oauth'
-                      ? 'border-purple-500 bg-purple-50 dark:bg-purple-900/30'
-                      : 'border-gray-200 dark:border-gray-700'
-                      }`}
-                    disabled={true}
-                    title="OAuth with Claude Pro/Max subscription is not yet available. Anthropic currently restricts third-party OAuth access."
-                  >
-                    <div class="font-medium text-sm text-gray-900 dark:text-gray-100 flex items-center gap-1.5">
-                      Claude Pro/Max
-                      <span class="inline-flex items-center px-1.5 py-0.5 text-[10px] font-semibold bg-gray-100 dark:bg-gray-800 text-gray-600 dark:text-gray-400 rounded">
-                        Unavailable
-                      </span>
-                    </div>
-                    <div class="text-xs text-gray-500 dark:text-gray-400 mt-0.5">
-                      Use your subscription
-                    </div>
-                  </button>
-                </div>
-                <p class={formHelpText}>
-                  {form.authMethod === 'api_key'
-                    ? 'Pay-per-use API billing from console.anthropic.com'
-                    : 'Use your Claude Pro ($20/mo) or Max ($100+/mo) subscription'}
-                </p>
-              </div>
-            </Show>
-
-            {/* OAuth Login/Status - shown when Anthropic + OAuth selected */}
-            <Show when={isAnthropicWithOAuth()}>
-              <div class={`${formField} p-4 rounded-lg border ${settings()?.oauth_connected
-                ? 'bg-green-50 dark:bg-green-900/20 border-green-200 dark:border-green-800'
-                : oauthAuthUrl()
-                  ? 'bg-amber-50 dark:bg-amber-900/20 border-amber-200 dark:border-amber-800'
-                  : 'bg-blue-50 dark:bg-blue-900/20 border-blue-200 dark:border-blue-800'
-                }`}>
-                {/* Connected state */}
-                <Show when={settings()?.oauth_connected}>
-                  <div class="flex items-center justify-between">
-                    <div>
-                      <div class="flex items-center gap-2 text-sm font-medium text-green-800 dark:text-green-200">
-                        <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24">
-                          <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M5 13l4 4L19 7" />
-                        </svg>
-                        Connected to Claude with your subscription
-                      </div>
-                      <p class="text-xs text-green-700 dark:text-green-300 mt-1">
-                        AI requests use your Claude Pro/Max subscription limits instead of API billing.
-                      </p>
-                    </div>
-                    <button
-                      type="button"
-                      class="px-3 py-1.5 text-xs border border-red-300 dark:border-red-700 text-red-700 dark:text-red-300 rounded hover:bg-red-50 dark:hover:bg-red-900/30 disabled:opacity-50"
-                      onClick={handleDisconnectOAuth}
-                      disabled={disconnectingOAuth() || saving()}
-                    >
-                      {disconnectingOAuth() ? 'Disconnecting...' : 'Disconnect'}
-                    </button>
-                  </div>
-                </Show>
-
-                {/* OAuth flow in progress - show URL and code input */}
-                <Show when={!settings()?.oauth_connected && oauthAuthUrl()}>
-                  <div class="space-y-4">
-                    <div class="text-sm text-amber-800 dark:text-amber-200">
-                      <strong>Step 1:</strong> Click the link below to sign in with your Anthropic account:
-                    </div>
-                    <a
-                      href={oauthAuthUrl()!}
-                      target="_blank"
-                      rel="noopener noreferrer"
-                      class="block p-2 bg-white dark:bg-gray-800 rounded border border-amber-300 dark:border-amber-700 text-xs text-blue-600 dark:text-blue-400 hover:underline break-all"
-                    >
-                      {oauthAuthUrl()}
-                    </a>
-                    <div class="text-sm text-amber-800 dark:text-amber-200">
-                      <strong>Step 2:</strong> After signing in, you'll see a code. Paste it here:
-                    </div>
-                    <div class="flex gap-2">
-                      <input
-                        type="text"
-                        value={oauthCode()}
-                        onInput={(e) => setOAuthCode(e.currentTarget.value)}
-                        placeholder="Paste authorization code here..."
-                        class="flex-1 px-3 py-2 text-sm border border-amber-300 dark:border-amber-600 rounded-md bg-white dark:bg-gray-800 text-gray-900 dark:text-gray-100 placeholder-gray-500 dark:placeholder-gray-400"
-                      />
-                      <button
-                        type="button"
-                        class="px-4 py-2 bg-gradient-to-r from-purple-600 to-pink-600 text-white text-sm rounded-md hover:from-purple-700 hover:to-pink-700 disabled:opacity-50 disabled:cursor-not-allowed"
-                        onClick={handleExchangeCode}
-                        disabled={exchangingCode() || !oauthCode().trim()}
-                      >
-                        {exchangingCode() ? 'Connecting...' : 'Connect'}
-                      </button>
-                    </div>
-                    <button
-                      type="button"
-                      class="text-xs text-gray-500 hover:text-gray-700 dark:text-gray-400 dark:hover:text-gray-200"
-                      onClick={handleCancelOAuth}
-                    >
-                      Cancel
-                    </button>
-                  </div>
-                </Show>
-
-                {/* Initial state - show sign in button */}
-                <Show when={!settings()?.oauth_connected && !oauthAuthUrl()}>
-                  <div class="text-center">
-                    <div class="text-sm text-blue-800 dark:text-blue-200 mb-3">
-                      <strong>Use your Claude subscription</strong>
-                      <p class="text-xs mt-1 text-blue-700 dark:text-blue-300">
-                        Sign in with your Anthropic account to use your Pro/Max subscription for AI features instead of API billing.
-                      </p>
-                    </div>
-                    <button
-                      type="button"
-                      class="px-4 py-2 bg-gradient-to-r from-purple-600 to-pink-600 text-white rounded-md hover:from-purple-700 hover:to-pink-700 disabled:opacity-50 disabled:cursor-not-allowed flex items-center gap-2 mx-auto"
-                      onClick={handleStartOAuth}
-                      disabled={startingOAuth() || saving()}
-                    >
-                      <Show when={startingOAuth()}>
-                        <span class="h-4 w-4 border-2 border-white border-t-transparent rounded-full animate-spin" />
-                      </Show>
-                      <Show when={!startingOAuth()}>
-                        <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24">
-                          <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M11 16l-4-4m0 0l4-4m-4 4h14m-5 4v1a3 3 0 01-3 3H6a3 3 0 01-3-3V7a3 3 0 013-3h7a3 3 0 013 3v1" />
-                        </svg>
-                      </Show>
-                      {startingOAuth() ? 'Starting...' : 'Sign in with Claude'}
-                    </button>
-                  </div>
-                </Show>
-              </div>
-            </Show>
-
-            {/* API Key - not shown for Ollama or when using OAuth */}
-            <Show when={needsApiKey()}>
-              <div class={formField}>
-                <div class="flex items-center justify-between">
-                  <label class={labelClass('mb-0')}>API Key</label>
-                  <Show when={settings()?.api_key_set}>
-                    <button
-                      type="button"
-                      class="text-xs text-purple-600 hover:underline dark:text-purple-300"
-                      onClick={() => {
-                        if (!saving()) {
-                          setForm('apiKey', '');
-                          setForm('clearApiKey', true);
-                          notificationStore.info('API key will be cleared on save', 2500);
-                        }
-                      }}
-                      disabled={saving()}
-                    >
-                      Clear stored key
-                    </button>
-                  </Show>
-                </div>
                 <input
-                  type="password"
+                  type="text"
-                  value={form.apiKey}
+                  value={form.model}
-                  onInput={(event) => {
+                  onInput={(e) => setForm('model', e.currentTarget.value)}
-                    setForm('apiKey', event.currentTarget.value);
+                  placeholder={DEFAULT_MODELS[form.provider]}
-                    if (event.currentTarget.value.trim() !== '') {
-                      setForm('clearApiKey', false);
-                    }
-                  }}
-                  placeholder={
-                    settings()?.api_key_set
-                      ? '•••••••• (leave blank to keep existing)'
-                      : `Enter ${PROVIDER_NAMES[form.provider]} API key`
-                  }
                   class={controlClass()}
                   disabled={saving()}
                 />
-                <p class={formHelpText}>
+              }>
-                  {form.provider === 'anthropic'
+                <select
-                    ? 'Get your API key from console.anthropic.com'
+                  value={form.model}
-                    : form.provider === 'deepseek'
+                  onChange={(e) => setForm('model', e.currentTarget.value)}
-                      ? 'Get your API key from platform.deepseek.com'
+                  class={controlClass()}
-                      : 'Get your API key from platform.openai.com'}
+                  disabled={saving()}
-                </p>
+                >
-              </div>
+                  <Show when={!form.model || !availableModels().some(m => m.id === form.model)}>
-            </Show>
+                    <option value={form.model}>{form.model || 'Select a model...'}</option>
-
+                  </Show>
-            {/* Model */}
+                  <For each={Array.from(groupModelsByProvider(availableModels()).entries())}>
-            <div class={formField}>
+                    {([provider, models]) => (
-              <label class={labelClass()}>Model</label>
+                      <optgroup label={PROVIDER_DISPLAY_NAMES[provider] || provider}>
-              <input
+                        <For each={models}>
-                type="text"
+                          {(model) => (
-                value={form.model}
+                            <option value={model.id}>
-                onInput={(event) => setForm('model', event.currentTarget.value)}
+                              {model.name || model.id.split(':').pop()}
-                placeholder={DEFAULT_MODELS[form.provider]}
+                            </option>
-                class={controlClass()}
+                          )}
-                disabled={saving()}
+                        </For>
-              />
+                      </optgroup>
+                    )}
+                  </For>
+                </select>
+              </Show>
               <p class={formHelpText}>
-                {form.provider === 'anthropic'
+                Main model used when no specific override is set. {availableModels().length === 0 && 'Save API key and refresh to see available models.'}
-                  ? 'e.g., claude-opus-4-5-20251101, claude-sonnet-4-20250514'
-                  : form.provider === 'openai'
-                    ? 'e.g., gpt-4o, gpt-4-turbo'
-                    : form.provider === 'deepseek'
-                      ? 'e.g., deepseek-chat, deepseek-coder'
-                      : 'e.g., llama3, mixtral, codellama'}
               </p>
             </div>
 
-            {/* Base URL - shown for Ollama (required) and OpenAI (optional) */}
+            {/* Chat Model Override */}
-            <Show when={showBaseUrl()}>
+            <div class={formField}>
-              <div class={formField}>
+              <label class={labelClass()}>Chat Model (Interactive)</label>
-                <label class={labelClass()}>
+              <Show when={availableModels().length > 0} fallback={
-                  {form.provider === 'ollama' ? 'Ollama Server URL' : 'API Base URL (optional)'}
-                </label>
                 <input
-                  type="url"
+                  type="text"
-                  value={form.baseUrl}
+                  value={form.chatModel}
-                  onInput={(event) => setForm('baseUrl', event.currentTarget.value)}
+                  onInput={(e) => setForm('chatModel', e.currentTarget.value)}
-                  placeholder={
+                  placeholder="Leave empty to use default model"
-                    form.provider === 'ollama'
-                      ? 'http://localhost:11434'
-                      : form.provider === 'deepseek'
-                        ? 'https://api.deepseek.com/chat/completions'
-                        : 'https://api.openai.com/v1'
-                  }
                   class={controlClass()}
                   disabled={saving()}
                 />
-                <p class={formHelpText}>
+              }>
-                  {form.provider === 'ollama'
+                <select
-                    ? 'URL where your Ollama server is running'
+                  value={form.chatModel}
-                    : form.provider === 'deepseek'
+                  onChange={(e) => setForm('chatModel', e.currentTarget.value)}
-                      ? 'Custom endpoint (leave blank for default DeepSeek API)'
+                  class={controlClass()}
-                      : 'Custom endpoint for Azure OpenAI or compatible APIs'}
+                  disabled={saving()}
+                >
+                  <option value="">Use default model ({form.model || 'not set'})</option>
+                  <For each={Array.from(groupModelsByProvider(availableModels()).entries())}>
+                    {([provider, models]) => (
+                      <optgroup label={PROVIDER_DISPLAY_NAMES[provider] || provider}>
+                        <For each={models}>
+                          {(model) => (
+                            <option value={model.id}>
+                              {model.name || model.id.split(':').pop()}
+                            </option>
+                          )}
+                        </For>
+                      </optgroup>
+                    )}
+                  </For>
+                </select>
+              </Show>
+              <p class={formHelpText}>
+                Model for interactive AI chat. Use a more capable model for complex reasoning.
+              </p>
+            </div>
+
+            {/* Patrol Model Override */}
+            <div class={formField}>
+              <label class={labelClass()}>Patrol Model (Background)</label>
+              <Show when={availableModels().length > 0} fallback={
+                <input
+                  type="text"
+                  value={form.patrolModel}
+                  onInput={(e) => setForm('patrolModel', e.currentTarget.value)}
+                  placeholder="Leave empty to use default model"
+                  class={controlClass()}
+                  disabled={saving()}
+                />
+              }>
+                <select
+                  value={form.patrolModel}
+                  onChange={(e) => setForm('patrolModel', e.currentTarget.value)}
+                  class={controlClass()}
+                  disabled={saving()}
+                >
+                  <option value="">Use default model ({form.model || 'not set'})</option>
+                  <For each={Array.from(groupModelsByProvider(availableModels()).entries())}>
+                    {([provider, models]) => (
+                      <optgroup label={PROVIDER_DISPLAY_NAMES[provider] || provider}>
+                        <For each={models}>
+                          {(model) => (
+                            <option value={model.id}>
+                              {model.name || model.id.split(':').pop()}
+                            </option>
+                          )}
+                        </For>
+                      </optgroup>
+                    )}
+                  </For>
+                </select>
+              </Show>
+              <p class={formHelpText}>
+                Model for background patrol analysis. Use a cheaper/faster model to save tokens.
+              </p>
+            </div>
+
+            {/* AI Provider Configuration - Configure API keys for all providers */}
+            <div class={`${formField} p-4 rounded-lg border bg-gradient-to-br from-purple-50 to-indigo-50 dark:from-purple-900/20 dark:to-indigo-900/20 border-purple-200 dark:border-purple-800`}>
+              <div class="mb-3">
+                <h4 class="font-medium text-gray-900 dark:text-white flex items-center gap-2">
+                  <svg class="w-5 h-5 text-purple-600 dark:text-purple-400" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                    <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 11H5m14 0a2 2 0 012 2v6a2 2 0 01-2 2H5a2 2 0 01-2-2v-6a2 2 0 012-2m14 0V9a2 2 0 00-2-2M5 11V9a2 2 0 012-2m0 0V5a2 2 0 012-2h6a2 2 0 012 2v2M7 7h10" />
+                  </svg>
+                  AI Provider Configuration
+                </h4>
+                <p class="text-xs text-gray-600 dark:text-gray-400 mt-1">
+                  Configure API keys for each AI provider you want to use. Models from all configured providers will appear in the model selectors.
                 </p>
               </div>
-            </Show>
+
+              {/* Provider Accordions */}
+              <div class="space-y-2">
+                {/* Anthropic */}
+                <div class={`border rounded-lg overflow-hidden ${settings()?.anthropic_configured ? 'border-green-300 dark:border-green-700' : 'border-gray-200 dark:border-gray-700'}`}>
+                  <button
+                    type="button"
+                    class="w-full px-3 py-2 flex items-center justify-between bg-white dark:bg-gray-800 hover:bg-gray-50 dark:hover:bg-gray-700/50 transition-colors"
+                    onClick={() => {
+                      const current = expandedProviders();
+                      const next = new Set(current);
+                      if (next.has('anthropic')) next.delete('anthropic');
+                      else next.add('anthropic');
+                      setExpandedProviders(next);
+                    }}
+                  >
+                    <div class="flex items-center gap-2">
+                      <span class="font-medium text-sm">Anthropic (Claude)</span>
+                      <Show when={settings()?.anthropic_configured}>
+                        <span class="px-1.5 py-0.5 text-[10px] font-semibold bg-green-100 dark:bg-green-900 text-green-700 dark:text-green-300 rounded">Configured</span>
+                      </Show>
+                    </div>
+                    <svg class={`w-4 h-4 transition-transform ${expandedProviders().has('anthropic') ? 'rotate-180' : ''}`} fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 9l-7 7-7-7" />
+                    </svg>
+                  </button>
+                  <Show when={expandedProviders().has('anthropic')}>
+                    <div class="px-3 py-3 bg-gray-50 dark:bg-gray-800/50 border-t border-gray-200 dark:border-gray-700">
+                      <input
+                        type="password"
+                        value={form.anthropicApiKey}
+                        onInput={(e) => setForm('anthropicApiKey', e.currentTarget.value)}
+                        placeholder={settings()?.anthropic_configured ? '••••••••••• (configured)' : 'sk-ant-... (from console.anthropic.com)'}
+                        class={controlClass()}
+                        disabled={saving()}
+                      />
+                      <p class="text-xs text-gray-500 mt-1">Get your API key from <a href="https://console.anthropic.com" target="_blank" class="text-purple-600 hover:underline">console.anthropic.com</a></p>
+                    </div>
+                  </Show>
+                </div>
+
+                {/* OpenAI */}
+                <div class={`border rounded-lg overflow-hidden ${settings()?.openai_configured ? 'border-green-300 dark:border-green-700' : 'border-gray-200 dark:border-gray-700'}`}>
+                  <button
+                    type="button"
+                    class="w-full px-3 py-2 flex items-center justify-between bg-white dark:bg-gray-800 hover:bg-gray-50 dark:hover:bg-gray-700/50 transition-colors"
+                    onClick={() => {
+                      const current = expandedProviders();
+                      const next = new Set(current);
+                      if (next.has('openai')) next.delete('openai');
+                      else next.add('openai');
+                      setExpandedProviders(next);
+                    }}
+                  >
+                    <div class="flex items-center gap-2">
+                      <span class="font-medium text-sm">OpenAI (GPT-4)</span>
+                      <Show when={settings()?.openai_configured}>
+                        <span class="px-1.5 py-0.5 text-[10px] font-semibold bg-green-100 dark:bg-green-900 text-green-700 dark:text-green-300 rounded">Configured</span>
+                      </Show>
+                    </div>
+                    <svg class={`w-4 h-4 transition-transform ${expandedProviders().has('openai') ? 'rotate-180' : ''}`} fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 9l-7 7-7-7" />
+                    </svg>
+                  </button>
+                  <Show when={expandedProviders().has('openai')}>
+                    <div class="px-3 py-3 bg-gray-50 dark:bg-gray-800/50 border-t border-gray-200 dark:border-gray-700 space-y-2">
+                      <input
+                        type="password"
+                        value={form.openaiApiKey}
+                        onInput={(e) => setForm('openaiApiKey', e.currentTarget.value)}
+                        placeholder={settings()?.openai_configured ? '••••••••••• (configured)' : 'sk-... (from platform.openai.com)'}
+                        class={controlClass()}
+                        disabled={saving()}
+                      />
+                      <input
+                        type="url"
+                        value={form.openaiBaseUrl}
+                        onInput={(e) => setForm('openaiBaseUrl', e.currentTarget.value)}
+                        placeholder="Custom base URL (optional, for Azure OpenAI)"
+                        class={controlClass()}
+                        disabled={saving()}
+                      />
+                      <p class="text-xs text-gray-500">Get your API key from <a href="https://platform.openai.com" target="_blank" class="text-purple-600 hover:underline">platform.openai.com</a></p>
+                    </div>
+                  </Show>
+                </div>
+
+                {/* DeepSeek */}
+                <div class={`border rounded-lg overflow-hidden ${settings()?.deepseek_configured ? 'border-green-300 dark:border-green-700' : 'border-gray-200 dark:border-gray-700'}`}>
+                  <button
+                    type="button"
+                    class="w-full px-3 py-2 flex items-center justify-between bg-white dark:bg-gray-800 hover:bg-gray-50 dark:hover:bg-gray-700/50 transition-colors"
+                    onClick={() => {
+                      const current = expandedProviders();
+                      const next = new Set(current);
+                      if (next.has('deepseek')) next.delete('deepseek');
+                      else next.add('deepseek');
+                      setExpandedProviders(next);
+                    }}
+                  >
+                    <div class="flex items-center gap-2">
+                      <span class="font-medium text-sm">DeepSeek</span>
+                      <Show when={settings()?.deepseek_configured}>
+                        <span class="px-1.5 py-0.5 text-[10px] font-semibold bg-green-100 dark:bg-green-900 text-green-700 dark:text-green-300 rounded">Configured</span>
+                      </Show>
+                    </div>
+                    <svg class={`w-4 h-4 transition-transform ${expandedProviders().has('deepseek') ? 'rotate-180' : ''}`} fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 9l-7 7-7-7" />
+                    </svg>
+                  </button>
+                  <Show when={expandedProviders().has('deepseek')}>
+                    <div class="px-3 py-3 bg-gray-50 dark:bg-gray-800/50 border-t border-gray-200 dark:border-gray-700">
+                      <input
+                        type="password"
+                        value={form.deepseekApiKey}
+                        onInput={(e) => setForm('deepseekApiKey', e.currentTarget.value)}
+                        placeholder={settings()?.deepseek_configured ? '••••••••••• (configured)' : 'sk-... (from platform.deepseek.com)'}
+                        class={controlClass()}
+                        disabled={saving()}
+                      />
+                      <p class="text-xs text-gray-500 mt-1">Get your API key from <a href="https://platform.deepseek.com" target="_blank" class="text-purple-600 hover:underline">platform.deepseek.com</a></p>
+                    </div>
+                  </Show>
+                </div>
+
+                {/* Ollama */}
+                <div class={`border rounded-lg overflow-hidden ${settings()?.ollama_configured ? 'border-green-300 dark:border-green-700' : 'border-gray-200 dark:border-gray-700'}`}>
+                  <button
+                    type="button"
+                    class="w-full px-3 py-2 flex items-center justify-between bg-white dark:bg-gray-800 hover:bg-gray-50 dark:hover:bg-gray-700/50 transition-colors"
+                    onClick={() => {
+                      const current = expandedProviders();
+                      const next = new Set(current);
+                      if (next.has('ollama')) next.delete('ollama');
+                      else next.add('ollama');
+                      setExpandedProviders(next);
+                    }}
+                  >
+                    <div class="flex items-center gap-2">
+                      <span class="font-medium text-sm">Ollama (Local)</span>
+                      <Show when={settings()?.ollama_configured}>
+                        <span class="px-1.5 py-0.5 text-[10px] font-semibold bg-green-100 dark:bg-green-900 text-green-700 dark:text-green-300 rounded">Available</span>
+                      </Show>
+                    </div>
+                    <svg class={`w-4 h-4 transition-transform ${expandedProviders().has('ollama') ? 'rotate-180' : ''}`} fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M19 9l-7 7-7-7" />
+                    </svg>
+                  </button>
+                  <Show when={expandedProviders().has('ollama')}>
+                    <div class="px-3 py-3 bg-gray-50 dark:bg-gray-800/50 border-t border-gray-200 dark:border-gray-700">
+                      <input
+                        type="url"
+                        value={form.ollamaBaseUrl}
+                        onInput={(e) => setForm('ollamaBaseUrl', e.currentTarget.value)}
+                        placeholder="http://localhost:11434"
+                        class={controlClass()}
+                        disabled={saving()}
+                      />
+                      <p class="text-xs text-gray-500 mt-1">URL where your Ollama server is running. No API key needed - it's free!</p>
+                    </div>
+                  </Show>
+                </div>
+              </div>
+            </div>
 
             {/* Autonomous Mode */}
             <div class={`${formField} p-4 rounded-lg border ${form.autonomousMode ? 'bg-amber-50 dark:bg-amber-900/20 border-amber-200 dark:border-amber-800' : 'bg-gray-50 dark:bg-gray-800/50 border-gray-200 dark:border-gray-700'}`}>
@@ -754,11 +816,7 @@ export const AISettings: Component = () => {
                   ? settings()?.oauth_connected
                     ? `Ready to use with ${settings()?.model} (via Claude subscription)`
                     : `Ready to use with ${settings()?.model}`
-                  : form.authMethod === 'oauth' && form.provider === 'anthropic'
+                  : 'Configure at least one AI provider above to enable AI features'}
-                    ? 'Sign in with your Claude subscription to enable AI features'
-                    : needsApiKey()
-                      ? 'API key required to enable AI features'
-                      : 'Configure Ollama server URL to enable AI features'}
               </span>
             </div>
           </Show>

frontend-modern/src/pages/Alerts.tsx

@@ -2171,12 +2171,12 @@ function OverviewTab(props: {
       }
 
       setPatrolStatus(status);
-      setAiFindings(findings);
+      setAiFindings(findings || []);
-      setPatrolRunHistory(runHistory);
+      setPatrolRunHistory(runHistory || []);
-      setSuppressionRules(rules);
+      setSuppressionRules(rules || []);
 
       // Auto-expand history if most recent run found issues
-      if (runHistory.length > 0 && runHistory[0].status !== 'healthy') {
+      if (runHistory && runHistory.length > 0 && runHistory[0].status !== 'healthy') {
         setShowRunHistory(true);
       }
     } catch (e) {
@@ -2624,7 +2624,8 @@ function OverviewTab(props: {
                                 onClick={(e) => {
                                   e.stopPropagation();
                                   aiChatStore.openWithPrompt(
-                                    `Help me fix this issue on ${finding.resource_name}: ${finding.title}\n\nDescription: ${finding.description}\n\nSuggested fix: ${finding.recommendation || 'None provided'}\n\nPlease guide me through applying this fix.`
+                                    `Help me fix this issue on ${finding.resource_name}: ${finding.title}\n\nDescription: ${finding.description}\n\nSuggested fix: ${finding.recommendation || 'None provided'}\n\nPlease guide me through applying this fix. When you've successfully fixed the issue, use the resolve_finding tool to mark it as resolved.`,
+                                    { findingId: finding.id }
                                   );
                                 }}
                               >

frontend-modern/src/stores/aiChat.ts

@@ -5,6 +5,7 @@ interface AIChatContext {
   targetId?: string;
   context?: Record<string, unknown>;
   initialPrompt?: string;
+  findingId?: string; // If opened from AI Insights "Get Help", the finding ID to resolve on success
 }
 
 // A single context item that can be accumulated
@@ -170,16 +171,16 @@ export const aiChatStore = {
   setTargetContext(targetType: string, targetId: string, additionalContext?: Record<string, unknown>) {
     // Use addContextItem instead of replacing
     const name = (additionalContext?.guestName as string) ||
-                 (additionalContext?.name as string) ||
+      (additionalContext?.name as string) ||
-                 targetId;
+      targetId;
     this.addContextItem(targetType, targetId, name, additionalContext || {});
   },
 
   // Open for a specific target - opens the panel and adds to context
   openForTarget(targetType: string, targetId: string, additionalContext?: Record<string, unknown>) {
     const name = (additionalContext?.guestName as string) ||
-                 (additionalContext?.name as string) ||
+      (additionalContext?.name as string) ||
-                 targetId;
+      targetId;
     this.addContextItem(targetType, targetId, name, additionalContext || {});
     setIsAIChatOpen(true);
   },

frontend-modern/src/types/ai.ts

@@ -3,12 +3,21 @@
 export type AIProvider = 'anthropic' | 'openai' | 'ollama' | 'deepseek';
 export type AuthMethod = 'api_key' | 'oauth';
 
+export interface ModelInfo {
+  id: string;
+  name: string;
+  description?: string;
+  is_default?: boolean;
+}
+
 export interface AISettings {
   enabled: boolean;
-  provider: AIProvider;
+  provider: AIProvider; // DEPRECATED: legacy single provider
-  api_key_set: boolean; // API key is never exposed, just whether it's set
+  api_key_set: boolean; // DEPRECATED: whether legacy API key is set
   model: string;
-  base_url?: string;
+  chat_model?: string; // Model for interactive chat (empty = use default)
+  patrol_model?: string; // Model for background patrol (empty = use default)
+  base_url?: string; // DEPRECATED: legacy base URL
   configured: boolean; // true if AI is ready to use
   autonomous_mode: boolean; // true if AI can execute commands without approval
   custom_context: string; // user-provided infrastructure context
@@ -19,21 +28,39 @@ export interface AISettings {
   patrol_schedule_preset?: string; // "15min" | "1hr" | "6hr" | "12hr" | "daily" | "disabled"
   alert_triggered_analysis?: boolean; // true if AI should analyze when alerts fire
   patrol_auto_fix?: boolean; // true if patrol can attempt automatic remediation
+  available_models?: ModelInfo[]; // DEPRECATED: use /api/ai/models endpoint
+  // Multi-provider configuration
+  anthropic_configured: boolean; // true if Anthropic API key or OAuth is set
+  openai_configured: boolean; // true if OpenAI API key is set
+  deepseek_configured: boolean; // true if DeepSeek API key is set
+  ollama_configured: boolean; // true (always available for attempt)
+  ollama_base_url: string; // Ollama server URL
+  openai_base_url?: string; // Custom OpenAI base URL
+  configured_providers: AIProvider[]; // List of providers with credentials
 }
 
 export interface AISettingsUpdateRequest {
   enabled?: boolean;
-  provider?: AIProvider;
+  provider?: AIProvider; // DEPRECATED: use model selection instead
-  api_key?: string; // empty string clears, undefined preserves
+  api_key?: string; // DEPRECATED: use per-provider keys
   model?: string;
-  base_url?: string;
+  base_url?: string; // DEPRECATED: use per-provider URLs
   autonomous_mode?: boolean;
   custom_context?: string; // user-provided infrastructure context
   auth_method?: AuthMethod; // "api_key" or "oauth"
+  // Model overrides for different use cases
+  chat_model?: string; // Model for interactive chat
+  patrol_model?: string; // Model for background patrol
   // Patrol settings for token efficiency
   patrol_schedule_preset?: string; // "15min" | "1hr" | "6hr" | "12hr" | "daily" | "disabled"
   alert_triggered_analysis?: boolean; // true if AI should analyze when alerts fire
   patrol_auto_fix?: boolean; // true if patrol can attempt automatic remediation
+  // Multi-provider credentials
+  anthropic_api_key?: string; // Set Anthropic API key
+  openai_api_key?: string; // Set OpenAI API key
+  deepseek_api_key?: string; // Set DeepSeek API key
+  ollama_base_url?: string; // Set Ollama server URL
+  openai_base_url?: string; // Set custom OpenAI base URL
 }
 
 
@@ -80,6 +107,8 @@ export interface AIExecuteRequest {
   target_id?: string;
   context?: Record<string, unknown>;
   history?: AIConversationMessage[]; // Previous conversation messages
+  finding_id?: string; // If fixing a patrol finding, the ID to resolve on success
+  model?: string; // Override model for this request (user selection in chat)
 }
 
 // Tool execution info

internal/ai/patrol.go

@@ -467,8 +467,7 @@ func (p *PatrolService) Start(ctx context.Context) {
 	p.mu.Unlock()
 
 	log.Info().
-		Dur("quick_interval", p.config.QuickCheckInterval).
+		Dur("interval", p.config.GetInterval()).
-		Dur("deep_interval", p.config.DeepAnalysisInterval).
 		Msg("Starting AI Patrol Service")
 
 	go p.patrolLoop(ctx)
@@ -1053,6 +1052,35 @@ func (p *PatrolService) GetFindingsSummary() FindingsSummary {
 	return p.findings.GetSummary()
 }
 
+// ResolveFinding marks a finding as resolved with a resolution note
+// This is called when the AI successfully fixes an issue
+func (p *PatrolService) ResolveFinding(findingID string, resolutionNote string) error {
+	if findingID == "" {
+		return fmt.Errorf("finding ID is required")
+	}
+
+	// Get the finding first to update its resolution note
+	finding := p.findings.Get(findingID)
+	if finding == nil {
+		return fmt.Errorf("finding not found: %s", findingID)
+	}
+
+	// Update the user note with the resolution
+	finding.UserNote = resolutionNote
+
+	// Mark as resolved (not auto-resolved since user/AI initiated it)
+	if !p.findings.Resolve(findingID, false) {
+		return fmt.Errorf("failed to resolve finding: %s", findingID)
+	}
+
+	log.Info().
+		Str("finding_id", findingID).
+		Str("resolution_note", resolutionNote).
+		Msg("AI resolved finding")
+
+	return nil
+}
+
 // GetRunHistory returns the history of patrol runs
 // If limit is > 0, returns at most that many records
 func (p *PatrolService) GetRunHistory(limit int) []PatrolRunRecord {
@@ -1419,7 +1447,7 @@ func (p *PatrolService) runAIAnalysis(ctx context.Context, state models.StateSna
 		return nil, nil // Nothing to analyze
 	}
 
-	prompt := p.buildPatrolPrompt(summary, false)
+	prompt := p.buildPatrolPrompt(summary)
 
 	log.Debug().Msg("AI Patrol: Sending infrastructure to LLM for analysis")
 
@@ -1434,6 +1462,7 @@ func (p *PatrolService) runAIAnalysis(ctx context.Context, state models.StateSna
 	resp, err := p.aiService.ExecuteStream(ctx, ExecuteRequest{
 		Prompt:       prompt,
 		SystemPrompt: p.getPatrolSystemPrompt(),
+		UseCase:      "patrol", // Use patrol model for background analysis
 	}, func(event StreamEvent) {
 		switch event.Type {
 		case "content":
@@ -1638,7 +1667,7 @@ func (p *PatrolService) buildInfrastructureSummary(state models.StateSnapshot) s
 
 // buildPatrolPrompt creates the prompt for AI analysis
 // Includes user feedback context to prevent re-raising dismissed findings
-func (p *PatrolService) buildPatrolPrompt(summary string, deep bool) string {
+func (p *PatrolService) buildPatrolPrompt(summary string) string {
 	// Get user feedback context (dismissed/snoozed findings)
 	feedbackContext := p.findings.GetDismissedForContext()
 	

internal/ai/providers/anthropic.go

@@ -346,3 +346,47 @@ func (c *AnthropicClient) TestConnection(ctx context.Context) error {
 	})
 	return err
 }
+
+// ListModels fetches available models from the Anthropic API
+func (c *AnthropicClient) ListModels(ctx context.Context) ([]ModelInfo, error) {
+	req, err := http.NewRequestWithContext(ctx, "GET", "https://api.anthropic.com/v1/models", nil)
+	if err != nil {
+		return nil, fmt.Errorf("failed to create request: %w", err)
+	}
+
+	req.Header.Set("x-api-key", c.apiKey)
+	req.Header.Set("anthropic-version", anthropicAPIVersion)
+
+	resp, err := c.client.Do(req)
+	if err != nil {
+		return nil, fmt.Errorf("request failed: %w", err)
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		body, _ := io.ReadAll(resp.Body)
+		return nil, fmt.Errorf("API error (%d): %s", resp.StatusCode, string(body))
+	}
+
+	var result struct {
+		Data []struct {
+			ID          string `json:"id"`
+			DisplayName string `json:"display_name"`
+			CreatedAt   string `json:"created_at"`
+		} `json:"data"`
+	}
+
+	if err := json.NewDecoder(resp.Body).Decode(&result); err != nil {
+		return nil, fmt.Errorf("failed to parse response: %w", err)
+	}
+
+	models := make([]ModelInfo, 0, len(result.Data))
+	for _, m := range result.Data {
+		models = append(models, ModelInfo{
+			ID:   m.ID,
+			Name: m.DisplayName,
+		})
+	}
+
+	return models, nil
+}

internal/ai/providers/anthropic_oauth.go

@@ -609,3 +609,52 @@ func (c *AnthropicOAuthClient) TestConnection(ctx context.Context) error {
 	})
 	return err
 }
+
+// ListModels fetches available models from the Anthropic API using OAuth
+func (c *AnthropicOAuthClient) ListModels(ctx context.Context) ([]ModelInfo, error) {
+	// Ensure we have a valid token
+	if err := c.ensureValidToken(ctx); err != nil {
+		return nil, err
+	}
+
+	req, err := http.NewRequestWithContext(ctx, "GET", "https://api.anthropic.com/v1/models", nil)
+	if err != nil {
+		return nil, fmt.Errorf("failed to create request: %w", err)
+	}
+
+	req.Header.Set("Authorization", "Bearer "+c.accessToken)
+	req.Header.Set("anthropic-version", anthropicAPIVersion)
+
+	resp, err := c.client.Do(req)
+	if err != nil {
+		return nil, fmt.Errorf("request failed: %w", err)
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		body, _ := io.ReadAll(resp.Body)
+		return nil, fmt.Errorf("API error (%d): %s", resp.StatusCode, string(body))
+	}
+
+	var result struct {
+		Data []struct {
+			ID          string `json:"id"`
+			DisplayName string `json:"display_name"`
+			CreatedAt   string `json:"created_at"`
+		} `json:"data"`
+	}
+
+	if err := json.NewDecoder(resp.Body).Decode(&result); err != nil {
+		return nil, fmt.Errorf("failed to parse response: %w", err)
+	}
+
+	models := make([]ModelInfo, 0, len(result.Data))
+	for _, m := range result.Data {
+		models = append(models, ModelInfo{
+			ID:   m.ID,
+			Name: m.DisplayName,
+		})
+	}
+
+	return models, nil
+}

internal/ai/providers/factory.go

@@ -7,6 +7,7 @@ import (
 )
 
 // NewFromConfig creates a Provider based on the AIConfig settings
+// DEPRECATED: Use NewForModel or NewForProvider for multi-provider support
 func NewFromConfig(cfg *config.AIConfig) (Provider, error) {
 	if cfg == nil {
 		return nil, fmt.Errorf("AI config is nil")
@@ -16,6 +17,13 @@ func NewFromConfig(cfg *config.AIConfig) (Provider, error) {
 		return nil, fmt.Errorf("AI is not enabled")
 	}
 
+	// Try multi-provider format first (uses per-provider API keys)
+	provider, model := config.ParseModelString(cfg.Model)
+	if providerClient, err := NewForProvider(cfg, provider, model); err == nil {
+		return providerClient, nil
+	}
+
+	// Fall back to legacy single-provider format
 	switch cfg.Provider {
 	case config.AIProviderAnthropic:
 		// If we have an API key (from direct entry or OAuth-created), use regular client
@@ -55,3 +63,57 @@ func NewFromConfig(cfg *config.AIConfig) (Provider, error) {
 	}
 }
 
+// NewForProvider creates a Provider for a specific provider using multi-provider credentials
+func NewForProvider(cfg *config.AIConfig, provider, model string) (Provider, error) {
+	if cfg == nil {
+		return nil, fmt.Errorf("AI config is nil")
+	}
+
+	switch provider {
+	case config.AIProviderAnthropic:
+		// Check for OAuth first
+		if cfg.IsUsingOAuth() && cfg.OAuthAccessToken != "" {
+			return NewAnthropicOAuthClient(
+				cfg.OAuthAccessToken,
+				cfg.OAuthRefreshToken,
+				cfg.OAuthExpiresAt,
+				model,
+			), nil
+		}
+		// Then check for per-provider API key
+		apiKey := cfg.GetAPIKeyForProvider(config.AIProviderAnthropic)
+		if apiKey == "" {
+			return nil, fmt.Errorf("Anthropic API key not configured")
+		}
+		return NewAnthropicClient(apiKey, model), nil
+
+	case config.AIProviderOpenAI:
+		apiKey := cfg.GetAPIKeyForProvider(config.AIProviderOpenAI)
+		if apiKey == "" {
+			return nil, fmt.Errorf("OpenAI API key not configured")
+		}
+		baseURL := cfg.GetBaseURLForProvider(config.AIProviderOpenAI)
+		return NewOpenAIClient(apiKey, model, baseURL), nil
+
+	case config.AIProviderDeepSeek:
+		apiKey := cfg.GetAPIKeyForProvider(config.AIProviderDeepSeek)
+		if apiKey == "" {
+			return nil, fmt.Errorf("DeepSeek API key not configured")
+		}
+		baseURL := cfg.GetBaseURLForProvider(config.AIProviderDeepSeek)
+		return NewOpenAIClient(apiKey, model, baseURL), nil
+
+	case config.AIProviderOllama:
+		baseURL := cfg.GetBaseURLForProvider(config.AIProviderOllama)
+		return NewOllamaClient(model, baseURL), nil
+
+	default:
+		return nil, fmt.Errorf("unknown provider: %s", provider)
+	}
+}
+
+// NewForModel creates a Provider for a specific model, automatically detecting the provider
+func NewForModel(cfg *config.AIConfig, modelString string) (Provider, error) {
+	provider, model := config.ParseModelString(modelString)
+	return NewForProvider(cfg, provider, model)
+}

internal/ai/providers/ollama.go

@@ -171,3 +171,45 @@ func (c *OllamaClient) TestConnection(ctx context.Context) error {
 
 	return nil
 }
+
+// ListModels fetches available models from the local Ollama instance
+func (c *OllamaClient) ListModels(ctx context.Context) ([]ModelInfo, error) {
+	url := c.baseURL + "/api/tags"
+	httpReq, err := http.NewRequestWithContext(ctx, "GET", url, nil)
+	if err != nil {
+		return nil, fmt.Errorf("failed to create request: %w", err)
+	}
+
+	resp, err := c.client.Do(httpReq)
+	if err != nil {
+		return nil, fmt.Errorf("request failed: %w", err)
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		body, _ := io.ReadAll(resp.Body)
+		return nil, fmt.Errorf("API error (%d): %s", resp.StatusCode, string(body))
+	}
+
+	var result struct {
+		Models []struct {
+			Name       string `json:"name"`
+			ModifiedAt string `json:"modified_at"`
+			Size       int64  `json:"size"`
+		} `json:"models"`
+	}
+
+	if err := json.NewDecoder(resp.Body).Decode(&result); err != nil {
+		return nil, fmt.Errorf("failed to parse response: %w", err)
+	}
+
+	models := make([]ModelInfo, 0, len(result.Models))
+	for _, m := range result.Models {
+		models = append(models, ModelInfo{
+			ID:   m.Name,
+			Name: m.Name,
+		})
+	}
+
+	return models, nil
+}

internal/ai/providers/openai.go

@@ -386,3 +386,58 @@ func (c *OpenAIClient) TestConnection(ctx context.Context) error {
 	})
 	return err
 }
+
+// ListModels fetches available models from the OpenAI API
+func (c *OpenAIClient) ListModels(ctx context.Context) ([]ModelInfo, error) {
+	// Determine the models endpoint based on the base URL
+	modelsURL := "https://api.openai.com/v1/models"
+	if c.isDeepSeek() {
+		modelsURL = "https://api.deepseek.com/models"
+	}
+
+	req, err := http.NewRequestWithContext(ctx, "GET", modelsURL, nil)
+	if err != nil {
+		return nil, fmt.Errorf("failed to create request: %w", err)
+	}
+
+	req.Header.Set("Authorization", "Bearer "+c.apiKey)
+
+	resp, err := c.client.Do(req)
+	if err != nil {
+		return nil, fmt.Errorf("request failed: %w", err)
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		body, _ := io.ReadAll(resp.Body)
+		return nil, fmt.Errorf("API error (%d): %s", resp.StatusCode, string(body))
+	}
+
+	var result struct {
+		Data []struct {
+			ID      string `json:"id"`
+			Object  string `json:"object"`
+			Created int64  `json:"created"`
+			OwnedBy string `json:"owned_by"`
+		} `json:"data"`
+	}
+
+	if err := json.NewDecoder(resp.Body).Decode(&result); err != nil {
+		return nil, fmt.Errorf("failed to parse response: %w", err)
+	}
+
+	models := make([]ModelInfo, 0, len(result.Data))
+	for _, m := range result.Data {
+		// Filter to only chat-capable models
+		if strings.Contains(m.ID, "gpt") || strings.Contains(m.ID, "o1") ||
+			strings.Contains(m.ID, "o3") || strings.Contains(m.ID, "deepseek") {
+			models = append(models, ModelInfo{
+				ID:        m.ID,
+				Name:      m.ID, // OpenAI uses ID as name
+				CreatedAt: m.Created,
+			})
+		}
+	}
+
+	return models, nil
+}

internal/ai/providers/provider.go

@@ -58,6 +58,14 @@ type ChatResponse struct {
 	OutputTokens     int         `json:"output_tokens,omitempty"`
 }
 
+// ModelInfo represents information about an available model
+type ModelInfo struct {
+	ID          string `json:"id"`
+	Name        string `json:"name"`
+	Description string `json:"description,omitempty"`
+	CreatedAt   int64  `json:"created_at,omitempty"`
+}
+
 // Provider defines the interface for AI providers
 type Provider interface {
 	// Chat sends a chat request and returns the response
@@ -68,4 +76,8 @@ type Provider interface {
 
 	// Name returns the provider name
 	Name() string
+
+	// ListModels returns available models from the provider's API
+	// Returns nil if the provider doesn't support listing models
+	ListModels(ctx context.Context) ([]ModelInfo, error)
 }

internal/ai/service.go

@@ -710,6 +710,9 @@ type ExecuteRequest struct {
 	Context      map[string]interface{} `json:"context,omitempty"`       // Current metrics, state, etc.
 	SystemPrompt string                 `json:"system_prompt,omitempty"` // Override system prompt
 	History      []ConversationMessage  `json:"history,omitempty"`       // Previous conversation messages
+	FindingID    string                 `json:"finding_id,omitempty"`    // If fixing a patrol finding, the ID to resolve
+	Model        string                 `json:"model,omitempty"`         // Override model for this request (for user selection in chat)
+	UseCase      string                 `json:"use_case,omitempty"`      // "chat" or "patrol" - determines which default model to use
 }
 
 // ExecuteResponse represents the AI's response
@@ -729,6 +732,34 @@ type ToolExecution struct {
 	Success bool   `json:"success"`
 }
 
+// getModelForRequest determines which model to use for a request
+// Priority: explicit override > use case default > config default
+func (s *Service) getModelForRequest(req ExecuteRequest) string {
+	// If request has explicit model override, use it
+	if req.Model != "" {
+		return req.Model
+	}
+
+	s.mu.RLock()
+	cfg := s.cfg
+	s.mu.RUnlock()
+
+	if cfg == nil {
+		return ""
+	}
+
+	// Use case-specific model selection
+	switch req.UseCase {
+	case "patrol":
+		return cfg.GetPatrolModel()
+	case "chat":
+		return cfg.GetChatModel()
+	default:
+		// Default to chat model for interactive requests
+		return cfg.GetChatModel()
+	}
+}
+
 // StreamEvent represents an event during AI execution for streaming
 type StreamEvent struct {
 	Type string      `json:"type"` // "tool_start", "tool_end", "content", "done", "error", "approval_needed"
@@ -765,11 +796,22 @@ type ApprovalNeededData struct {
 // If tools are available and the AI requests them, it executes them in a loop
 func (s *Service) Execute(ctx context.Context, req ExecuteRequest) (*ExecuteResponse, error) {
 	s.mu.RLock()
-	provider := s.provider
+	defaultProvider := s.provider
-	cfg := s.cfg
 	agentServer := s.agentServer
+	cfg := s.cfg
 	s.mu.RUnlock()
 
+	// Determine the model to use for this request
+	modelString := s.getModelForRequest(req)
+
+	// Create a provider for this specific model (supports multi-provider switching)
+	provider, err := providers.NewForModel(cfg, modelString)
+	if err != nil {
+		// Fall back to default provider if model-specific provider can't be created
+		log.Debug().Err(err).Str("model", modelString).Msg("Could not create provider for model, using default")
+		provider = defaultProvider
+	}
+
 	if provider == nil {
 		return nil, fmt.Errorf("AI is not enabled or configured")
 	}
@@ -836,7 +878,7 @@ Always execute the commands rather than telling the user how to do it.`
 	for i := 0; i < maxIterations; i++ {
 		resp, err := provider.Chat(ctx, providers.ChatRequest{
 			Messages:  messages,
-			Model:     cfg.GetModel(),
+			Model:     s.getModelForRequest(req),
 			System:    systemPrompt,
 			MaxTokens: 4096,
 			Tools:     tools,
@@ -893,11 +935,22 @@ Always execute the commands rather than telling the user how to do it.`
 // This allows the UI to show real-time progress during tool execution
 func (s *Service) ExecuteStream(ctx context.Context, req ExecuteRequest, callback StreamCallback) (*ExecuteResponse, error) {
 	s.mu.RLock()
-	provider := s.provider
+	defaultProvider := s.provider
-	cfg := s.cfg
 	agentServer := s.agentServer
+	cfg := s.cfg
 	s.mu.RUnlock()
 
+	// Determine the model to use for this request
+	modelString := s.getModelForRequest(req)
+
+	// Create a provider for this specific model (supports multi-provider switching)
+	provider, err := providers.NewForModel(cfg, modelString)
+	if err != nil {
+		// Fall back to default provider if model-specific provider can't be created
+		log.Debug().Err(err).Str("model", modelString).Msg("Could not create provider for model, using default")
+		provider = defaultProvider
+	}
+
 	if provider == nil {
 		return nil, fmt.Errorf("AI is not enabled or configured")
 	}
@@ -994,7 +1047,7 @@ Always execute the commands rather than telling the user how to do it.`
 
 		resp, err := provider.Chat(ctx, providers.ChatRequest{
 			Messages:  messages,
-			Model:     cfg.GetModel(),
+			Model:     s.getModelForRequest(req),
 			System:    systemPrompt,
 			MaxTokens: 4096,
 			Tools:     tools,
@@ -1327,6 +1380,24 @@ func (s *Service) getTools() []providers.Tool {
 				"required": []string{"resource_type", "resource_id", "url"},
 			},
 		},
+		{
+			Name:        "resolve_finding",
+			Description: "Mark an AI patrol finding as resolved after you have successfully fixed the underlying issue. Only use this after confirming the fix worked (e.g., by running a verification command). The finding ID is provided in your context when helping with a patrol finding.",
+			InputSchema: map[string]interface{}{
+				"type": "object",
+				"properties": map[string]interface{}{
+					"finding_id": map[string]interface{}{
+						"type":        "string",
+						"description": "The ID of the finding to resolve. Use the finding_id from the request context.",
+					},
+					"resolution_note": map[string]interface{}{
+						"type":        "string",
+						"description": "Brief description of how the issue was resolved (e.g., 'Restarted nginx service', 'Cleaned up disk space').",
+					},
+				},
+				"required": []string{"finding_id", "resolution_note"},
+			},
+		},
 	}
 
 	// Add web search tool for Anthropic provider
@@ -1607,6 +1678,47 @@ func (s *Service) executeTool(ctx context.Context, req ExecuteRequest, tc provid
 		execution.Success = true
 		return execution.Output, execution
 
+	case "resolve_finding":
+		findingID, _ := tc.Input["finding_id"].(string)
+		resolutionNote, _ := tc.Input["resolution_note"].(string)
+		execution.Input = fmt.Sprintf("finding: %s, note: %s", findingID, resolutionNote)
+
+		// If no finding ID provided by AI, check the request context
+		if findingID == "" {
+			findingID = req.FindingID
+		}
+
+		if findingID == "" {
+			execution.Output = "Error: finding_id is required. The finding ID should be provided in the request context when helping fix a patrol finding."
+			return execution.Output, execution
+		}
+
+		if resolutionNote == "" {
+			execution.Output = "Error: resolution_note is required. Please describe how the issue was resolved."
+			return execution.Output, execution
+		}
+
+		// Get the patrol service to resolve the finding
+		s.mu.RLock()
+		patrolService := s.patrolService
+		s.mu.RUnlock()
+
+		if patrolService == nil {
+			execution.Output = "Error: Patrol service not available"
+			return execution.Output, execution
+		}
+
+		// Resolve the finding
+		err := patrolService.ResolveFinding(findingID, resolutionNote)
+		if err != nil {
+			execution.Output = fmt.Sprintf("Error resolving finding: %s", err)
+			return execution.Output, execution
+		}
+
+		execution.Output = fmt.Sprintf("✅ Finding resolved! The AI Insight has been marked as fixed.\nID: %s\nResolution: %s", findingID, resolutionNote)
+		execution.Success = true
+		return execution.Output, execution
+
 	default:
 		execution.Output = fmt.Sprintf("Unknown tool: %s", tc.Name)
 		return execution.Output, execution
@@ -2255,31 +2367,109 @@ func (s *Service) buildUserAnnotationsContext() string {
 }
 
 // TestConnection tests the AI provider connection
+// Tests the provider for the currently configured default model
 func (s *Service) TestConnection(ctx context.Context) error {
 	s.mu.RLock()
-	provider := s.provider
+	cfg := s.cfg
+	defaultProvider := s.provider
 	s.mu.RUnlock()
 
-	if provider == nil {
+	// Load config if not available
-		// Try to create a temporary provider from config to test
+	if cfg == nil {
-		cfg, err := s.persistence.LoadAIConfig()
+		var err error
+		cfg, err = s.persistence.LoadAIConfig()
 		if err != nil {
 			return fmt.Errorf("failed to load AI config: %w", err)
 		}
-		if cfg == nil || cfg.APIKey == "" {
+	}
-			return fmt.Errorf("API key not configured")
-		}
 
-		tempProvider, err := providers.NewFromConfig(cfg)
+	if cfg == nil || !cfg.IsConfigured() {
-		if err != nil {
+		return fmt.Errorf("no AI provider configured")
-			return err
+	}
+
+	// Try to create a provider for the current default model
+	provider, err := providers.NewForModel(cfg, cfg.GetModel())
+	if err != nil {
+		// Fall back to default provider or NewFromConfig
+		log.Debug().Err(err).Str("model", cfg.GetModel()).Msg("Could not create provider for model, using fallback")
+		if defaultProvider != nil {
+			provider = defaultProvider
+		} else {
+			provider, err = providers.NewFromConfig(cfg)
+			if err != nil {
+				return err
+			}
 		}
-		provider = tempProvider
 	}
 
 	return provider.TestConnection(ctx)
 }
 
+// ListModels fetches available models from ALL configured AI providers
+// Returns a unified list with models prefixed by provider name
+func (s *Service) ListModels(ctx context.Context) ([]providers.ModelInfo, error) {
+	cfg, err := s.persistence.LoadAIConfig()
+	if err != nil {
+		return nil, fmt.Errorf("failed to load AI config: %w", err)
+	}
+	if cfg == nil {
+		return nil, fmt.Errorf("AI not configured")
+	}
+
+	var allModels []providers.ModelInfo
+
+	// Query each configured provider
+	providersList := []string{config.AIProviderAnthropic, config.AIProviderOpenAI, config.AIProviderDeepSeek, config.AIProviderOllama}
+
+	for _, providerName := range providersList {
+		if !cfg.HasProvider(providerName) {
+			continue
+		}
+
+		// Create provider for this specific provider
+		provider, err := providers.NewForProvider(cfg, providerName, "")
+		if err != nil {
+			log.Debug().Err(err).Str("provider", providerName).Msg("Skipping provider - not configured")
+			continue
+		}
+
+		// Fetch models from this provider
+		models, err := provider.ListModels(ctx)
+		if err != nil {
+			log.Warn().Err(err).Str("provider", providerName).Msg("Failed to fetch models from provider")
+			continue
+		}
+
+		// Add provider prefix to each model
+		for _, m := range models {
+			allModels = append(allModels, providers.ModelInfo{
+				ID:          config.FormatModelString(providerName, m.ID),
+				Name:        m.Name,
+				Description: providerDisplayName(providerName) + ": " + m.ID,
+				CreatedAt:   m.CreatedAt,
+			})
+		}
+	}
+
+	return allModels, nil
+}
+
+// providerDisplayName returns a user-friendly name for a provider
+func providerDisplayName(provider string) string {
+	switch provider {
+	case config.AIProviderAnthropic:
+		return "Anthropic"
+	case config.AIProviderOpenAI:
+		return "OpenAI"
+	case config.AIProviderDeepSeek:
+		return "DeepSeek"
+	case config.AIProviderOllama:
+		return "Ollama"
+	default:
+		return provider
+	}
+}
+
 // Reload reloads the AI configuration (call after settings change)
 func (s *Service) Reload() error {
 	return s.LoadConfig()

internal/api/ai_handlers.go

@@ -102,13 +102,15 @@ func (h *AISettingsHandler) GetAlertTriggeredAnalyzer() *ai.AlertTriggeredAnalyz
 }
 
 // AISettingsResponse is returned by GET /api/settings/ai
-// API key is masked for security
+// API keys are masked for security
 type AISettingsResponse struct {
 	Enabled        bool   `json:"enabled"`
-	Provider       string `json:"provider"`
+	Provider       string `json:"provider"`          // DEPRECATED: legacy single provider
-	APIKeySet      bool   `json:"api_key_set"` // true if API key is configured (never expose actual key)
+	APIKeySet      bool   `json:"api_key_set"`       // DEPRECATED: true if legacy API key is configured
 	Model          string `json:"model"`
-	BaseURL        string `json:"base_url,omitempty"`
+	ChatModel      string `json:"chat_model,omitempty"`   // Model for interactive chat (empty = use default)
+	PatrolModel    string `json:"patrol_model,omitempty"` // Model for patrol (empty = use default)
+	BaseURL        string `json:"base_url,omitempty"`     // DEPRECATED: legacy base URL
 	Configured     bool   `json:"configured"`      // true if AI is ready to use
 	AutonomousMode bool   `json:"autonomous_mode"` // true if AI can execute without approval
 	CustomContext  string `json:"custom_context"`  // user-provided infrastructure context
@@ -116,23 +118,40 @@ type AISettingsResponse struct {
 	AuthMethod     string `json:"auth_method"`     // "api_key" or "oauth"
 	OAuthConnected bool   `json:"oauth_connected"` // true if OAuth tokens are configured
 	// Patrol settings for token efficiency
-	PatrolSchedulePreset   string `json:"patrol_schedule_preset"`   // "15min", "1hr", "6hr", "12hr", "daily", "disabled"
+	PatrolSchedulePreset   string              `json:"patrol_schedule_preset"`   // "15min", "1hr", "6hr", "12hr", "daily", "disabled"
-	AlertTriggeredAnalysis bool   `json:"alert_triggered_analysis"` // true if AI analyzes when alerts fire
+	AlertTriggeredAnalysis bool                `json:"alert_triggered_analysis"` // true if AI analyzes when alerts fire
+	AvailableModels        []config.ModelInfo  `json:"available_models"`         // List of models for current provider
+	// Multi-provider credentials - shows which providers are configured
+	AnthropicConfigured bool   `json:"anthropic_configured"` // true if Anthropic API key or OAuth is set
+	OpenAIConfigured    bool   `json:"openai_configured"`    // true if OpenAI API key is set
+	DeepSeekConfigured  bool   `json:"deepseek_configured"`  // true if DeepSeek API key is set
+	OllamaConfigured    bool   `json:"ollama_configured"`    // true (always available for attempt)
+	OllamaBaseURL       string `json:"ollama_base_url"`      // Ollama server URL
+	OpenAIBaseURL       string `json:"openai_base_url,omitempty"` // Custom OpenAI base URL
+	ConfiguredProviders []string `json:"configured_providers"` // List of provider names with credentials
 }
 
 // AISettingsUpdateRequest is the request body for PUT /api/settings/ai
 type AISettingsUpdateRequest struct {
 	Enabled        *bool   `json:"enabled,omitempty"`
-	Provider       *string `json:"provider,omitempty"`
+	Provider       *string `json:"provider,omitempty"`    // DEPRECATED: use model selection instead
-	APIKey         *string `json:"api_key,omitempty"` // empty string clears, null preserves
+	APIKey         *string `json:"api_key,omitempty"`     // DEPRECATED: use per-provider keys
 	Model          *string `json:"model,omitempty"`
-	BaseURL        *string `json:"base_url,omitempty"`
+	ChatModel      *string `json:"chat_model,omitempty"`   // Model for interactive chat
+	PatrolModel    *string `json:"patrol_model,omitempty"` // Model for background patrol
+	BaseURL        *string `json:"base_url,omitempty"`     // DEPRECATED: use per-provider URLs
 	AutonomousMode *bool   `json:"autonomous_mode,omitempty"`
 	CustomContext  *string `json:"custom_context,omitempty"` // user-provided infrastructure context
 	AuthMethod     *string `json:"auth_method,omitempty"`    // "api_key" or "oauth"
 	// Patrol settings for token efficiency
 	PatrolSchedulePreset   *string `json:"patrol_schedule_preset,omitempty"`   // "15min", "1hr", "6hr", "12hr", "daily", "disabled"
 	AlertTriggeredAnalysis *bool   `json:"alert_triggered_analysis,omitempty"` // true if AI analyzes when alerts fire
+	// Multi-provider credentials
+	AnthropicAPIKey *string `json:"anthropic_api_key,omitempty"` // Set Anthropic API key
+	OpenAIAPIKey    *string `json:"openai_api_key,omitempty"`    // Set OpenAI API key
+	DeepSeekAPIKey  *string `json:"deepseek_api_key,omitempty"`  // Set DeepSeek API key
+	OllamaBaseURL   *string `json:"ollama_base_url,omitempty"`   // Set Ollama server URL
+	OpenAIBaseURL   *string `json:"openai_base_url,omitempty"`   // Set custom OpenAI base URL
 }
 
 // HandleGetAISettings returns the current AI settings (GET /api/settings/ai)
@@ -164,6 +183,8 @@ func (h *AISettingsHandler) HandleGetAISettings(w http.ResponseWriter, r *http.R
 		Provider:       settings.Provider,
 		APIKeySet:      settings.APIKey != "",
 		Model:          settings.GetModel(),
+		ChatModel:      settings.ChatModel,
+		PatrolModel:    settings.PatrolModel,
 		BaseURL:        settings.BaseURL,
 		Configured:     settings.IsConfigured(),
 		AutonomousMode: settings.AutonomousMode,
@@ -173,6 +194,15 @@ func (h *AISettingsHandler) HandleGetAISettings(w http.ResponseWriter, r *http.R
 		// Patrol settings
 		PatrolSchedulePreset:   settings.PatrolSchedulePreset,
 		AlertTriggeredAnalysis: settings.AlertTriggeredAnalysis,
+		AvailableModels:        nil, // Now populated via /api/ai/models endpoint
+		// Multi-provider configuration
+		AnthropicConfigured: settings.HasProvider(config.AIProviderAnthropic),
+		OpenAIConfigured:    settings.HasProvider(config.AIProviderOpenAI),
+		DeepSeekConfigured:  settings.HasProvider(config.AIProviderDeepSeek),
+		OllamaConfigured:    settings.HasProvider(config.AIProviderOllama),
+		OllamaBaseURL:       settings.GetBaseURLForProvider(config.AIProviderOllama),
+		OpenAIBaseURL:       settings.OpenAIBaseURL,
+		ConfiguredProviders: settings.GetConfiguredProviders(),
 	}
 
 	if err := utils.WriteJSONResponse(w, response); err != nil {
@@ -247,6 +277,14 @@ func (h *AISettingsHandler) HandleUpdateAISettings(w http.ResponseWriter, r *htt
 		settings.Model = strings.TrimSpace(*req.Model)
 	}
 
+	if req.ChatModel != nil {
+		settings.ChatModel = strings.TrimSpace(*req.ChatModel)
+	}
+
+	if req.PatrolModel != nil {
+		settings.PatrolModel = strings.TrimSpace(*req.PatrolModel)
+	}
+
 	if req.BaseURL != nil {
 		settings.BaseURL = strings.TrimSpace(*req.BaseURL)
 	}
@@ -296,6 +334,23 @@ func (h *AISettingsHandler) HandleUpdateAISettings(w http.ResponseWriter, r *htt
 		settings.AlertTriggeredAnalysis = *req.AlertTriggeredAnalysis
 	}
 
+	// Handle multi-provider credentials
+	if req.AnthropicAPIKey != nil {
+		settings.AnthropicAPIKey = strings.TrimSpace(*req.AnthropicAPIKey)
+	}
+	if req.OpenAIAPIKey != nil {
+		settings.OpenAIAPIKey = strings.TrimSpace(*req.OpenAIAPIKey)
+	}
+	if req.DeepSeekAPIKey != nil {
+		settings.DeepSeekAPIKey = strings.TrimSpace(*req.DeepSeekAPIKey)
+	}
+	if req.OllamaBaseURL != nil {
+		settings.OllamaBaseURL = strings.TrimSpace(*req.OllamaBaseURL)
+	}
+	if req.OpenAIBaseURL != nil {
+		settings.OpenAIBaseURL = strings.TrimSpace(*req.OpenAIBaseURL)
+	}
+
 	// Save settings
 	if err := h.persistence.SaveAIConfig(*settings); err != nil {
 		log.Error().Err(err).Msg("Failed to save AI settings")
@@ -336,6 +391,8 @@ func (h *AISettingsHandler) HandleUpdateAISettings(w http.ResponseWriter, r *htt
 		Provider:               settings.Provider,
 		APIKeySet:              settings.APIKey != "",
 		Model:                  settings.GetModel(),
+		ChatModel:              settings.ChatModel,
+		PatrolModel:            settings.PatrolModel,
 		BaseURL:                settings.BaseURL,
 		Configured:             settings.IsConfigured(),
 		AutonomousMode:         settings.AutonomousMode,
@@ -344,6 +401,15 @@ func (h *AISettingsHandler) HandleUpdateAISettings(w http.ResponseWriter, r *htt
 		OAuthConnected:         settings.OAuthAccessToken != "",
 		PatrolSchedulePreset:   settings.PatrolSchedulePreset,
 		AlertTriggeredAnalysis: settings.AlertTriggeredAnalysis,
+		AvailableModels:        nil, // Now populated via /api/ai/models endpoint
+		// Multi-provider configuration
+		AnthropicConfigured: settings.HasProvider(config.AIProviderAnthropic),
+		OpenAIConfigured:    settings.HasProvider(config.AIProviderOpenAI),
+		DeepSeekConfigured:  settings.HasProvider(config.AIProviderDeepSeek),
+		OllamaConfigured:    settings.HasProvider(config.AIProviderOllama),
+		OllamaBaseURL:       settings.GetBaseURLForProvider(config.AIProviderOllama),
+		OpenAIBaseURL:       settings.OpenAIBaseURL,
+		ConfiguredProviders: settings.GetConfiguredProviders(),
 	}
 
 	if err := utils.WriteJSONResponse(w, response); err != nil {
@@ -390,6 +456,65 @@ func (h *AISettingsHandler) HandleTestAIConnection(w http.ResponseWriter, r *htt
 	}
 }
 
+// HandleListModels fetches available models from the configured AI provider (GET /api/ai/models)
+func (h *AISettingsHandler) HandleListModels(w http.ResponseWriter, r *http.Request) {
+	if r.Method != http.MethodGet {
+		http.Error(w, "Method not allowed", http.StatusMethodNotAllowed)
+		return
+	}
+
+	// Require authentication
+	if !CheckAuth(h.config, w, r) {
+		return
+	}
+
+	ctx, cancel := context.WithTimeout(r.Context(), 30*time.Second)
+	defer cancel()
+
+	type ModelInfo struct {
+		ID          string `json:"id"`
+		Name        string `json:"name"`
+		Description string `json:"description,omitempty"`
+	}
+
+	type Response struct {
+		Models  []ModelInfo `json:"models"`
+		Error   string      `json:"error,omitempty"`
+		Cached  bool        `json:"cached"`
+	}
+
+	models, err := h.aiService.ListModels(ctx)
+	if err != nil {
+		// Return error but don't fail the request - frontend can show a fallback
+		resp := Response{
+			Models: []ModelInfo{},
+			Error:  err.Error(),
+		}
+		if jsonErr := utils.WriteJSONResponse(w, resp); jsonErr != nil {
+			log.Error().Err(jsonErr).Msg("Failed to write AI models response")
+		}
+		return
+	}
+
+	// Convert provider models to response format
+	responseModels := make([]ModelInfo, 0, len(models))
+	for _, m := range models {
+		responseModels = append(responseModels, ModelInfo{
+			ID:          m.ID,
+			Name:        m.Name,
+			Description: m.Description,
+		})
+	}
+
+	resp := Response{
+		Models: responseModels,
+	}
+
+	if err := utils.WriteJSONResponse(w, resp); err != nil {
+		log.Error().Err(err).Msg("Failed to write AI models response")
+	}
+}
+
 // AIExecuteRequest is the request body for POST /api/ai/execute
 // AIConversationMessage represents a message in conversation history
 type AIConversationMessage struct {

internal/api/router.go

@@ -1076,6 +1076,7 @@ func (r *Router) setupRoutes() {
 	r.mux.HandleFunc("/api/settings/ai", RequireAdmin(r.config, RequireScope(config.ScopeSettingsRead, r.aiSettingsHandler.HandleGetAISettings)))
 	r.mux.HandleFunc("/api/settings/ai/update", RequireAdmin(r.config, RequireScope(config.ScopeSettingsWrite, r.aiSettingsHandler.HandleUpdateAISettings)))
 	r.mux.HandleFunc("/api/ai/test", RequireAdmin(r.config, RequireScope(config.ScopeSettingsWrite, r.aiSettingsHandler.HandleTestAIConnection)))
+	r.mux.HandleFunc("/api/ai/models", RequireAuth(r.config, r.aiSettingsHandler.HandleListModels))
 	r.mux.HandleFunc("/api/ai/execute", RequireAuth(r.config, r.aiSettingsHandler.HandleExecute))
 	r.mux.HandleFunc("/api/ai/execute/stream", RequireAuth(r.config, r.aiSettingsHandler.HandleExecuteStream))
 	r.mux.HandleFunc("/api/ai/investigate-alert", RequireAuth(r.config, r.aiSettingsHandler.HandleInvestigateAlert))

internal/config/ai.go

@@ -16,13 +16,22 @@ const (
 // This is stored in ai.enc (encrypted) in the config directory
 type AIConfig struct {
 	Enabled        bool   `json:"enabled"`
-	Provider       string `json:"provider"`        // "anthropic", "openai", "ollama", "deepseek"
+	Provider       string `json:"provider"`        // DEPRECATED: legacy single provider field, kept for migration
-	APIKey         string `json:"api_key"`         // encrypted at rest (not needed for ollama or oauth)
+	APIKey         string `json:"api_key"`         // DEPRECATED: legacy single API key, kept for migration
-	Model          string `json:"model"`           // e.g., "claude-opus-4-5-20250514", "gpt-4o", "llama3"
+	Model          string `json:"model"`           // Currently selected default model (format: "provider:model-name")
-	BaseURL        string `json:"base_url"`        // custom endpoint (required for ollama, optional for openai)
+	ChatModel      string `json:"chat_model,omitempty"`   // Model for interactive chat (defaults to Model)
+	PatrolModel    string `json:"patrol_model,omitempty"` // Model for background patrol (defaults to Model, can be cheaper)
+	BaseURL        string `json:"base_url"`        // DEPRECATED: legacy base URL, kept for migration
 	AutonomousMode bool   `json:"autonomous_mode"` // when true, AI executes commands without approval
 	CustomContext  string `json:"custom_context"`  // user-provided context about their infrastructure
 
+	// Multi-provider credentials - each provider can be configured independently
+	AnthropicAPIKey string `json:"anthropic_api_key,omitempty"` // Anthropic API key
+	OpenAIAPIKey    string `json:"openai_api_key,omitempty"`    // OpenAI API key
+	DeepSeekAPIKey  string `json:"deepseek_api_key,omitempty"`  // DeepSeek API key
+	OllamaBaseURL   string `json:"ollama_base_url,omitempty"`   // Ollama server URL (default: http://localhost:11434)
+	OpenAIBaseURL   string `json:"openai_base_url,omitempty"`   // Custom OpenAI-compatible base URL (optional)
+
 	// OAuth fields for Claude Pro/Max subscription authentication
 	AuthMethod        AuthMethod `json:"auth_method,omitempty"`         // "api_key" or "oauth" (for anthropic only)
 	OAuthAccessToken  string     `json:"oauth_access_token,omitempty"`  // OAuth access token (encrypted at rest)
@@ -61,6 +70,21 @@ const (
 	DefaultDeepSeekBaseURL  = "https://api.deepseek.com/chat/completions"
 )
 
+// ModelInfo represents information about an available model
+// Deprecated: Use providers.ModelInfo instead - models are now fetched dynamically from APIs
+type ModelInfo struct {
+	ID          string `json:"id"`
+	Name        string `json:"name"`
+	Description string `json:"description,omitempty"`
+	IsDefault   bool   `json:"is_default,omitempty"`
+}
+
+// GetAvailableModels is deprecated - models are now fetched dynamically from provider APIs
+// This returns nil; use the /api/ai/models endpoint instead which queries the actual API
+func GetAvailableModels(provider string) []ModelInfo {
+	return nil
+}
+
 // NewDefaultAIConfig returns an AIConfig with sensible defaults
 func NewDefaultAIConfig() *AIConfig {
 	return &AIConfig{
@@ -83,14 +107,21 @@ func NewDefaultAIConfig() *AIConfig {
 }
 
 // IsConfigured returns true if the AI config has enough info to make API calls
+// For multi-provider setup, returns true if ANY provider is configured
 func (c *AIConfig) IsConfigured() bool {
 	if !c.Enabled {
 		return false
 	}
 
+	// Check multi-provider credentials first (new format)
+	if c.HasProvider(AIProviderAnthropic) || c.HasProvider(AIProviderOpenAI) ||
+		c.HasProvider(AIProviderDeepSeek) || c.HasProvider(AIProviderOllama) {
+		return true
+	}
+
+	// Fall back to legacy single-provider check for backward compatibility
 	switch c.Provider {
 	case AIProviderAnthropic:
-		// Anthropic can use API key OR OAuth
 		if c.AuthMethod == AuthMethodOAuth {
 			return c.OAuthAccessToken != ""
 		}
@@ -98,19 +129,140 @@ func (c *AIConfig) IsConfigured() bool {
 	case AIProviderOpenAI, AIProviderDeepSeek:
 		return c.APIKey != ""
 	case AIProviderOllama:
-		// Ollama doesn't need an API key
 		return true
 	default:
 		return false
 	}
 }
 
+// HasProvider returns true if the specified provider has credentials configured
+func (c *AIConfig) HasProvider(provider string) bool {
+	switch provider {
+	case AIProviderAnthropic:
+		// Anthropic can use API key OR OAuth
+		if c.AuthMethod == AuthMethodOAuth && c.OAuthAccessToken != "" {
+			return true
+		}
+		return c.AnthropicAPIKey != ""
+	case AIProviderOpenAI:
+		return c.OpenAIAPIKey != ""
+	case AIProviderDeepSeek:
+		return c.DeepSeekAPIKey != ""
+	case AIProviderOllama:
+		// Ollama is available if URL is set (or default localhost)
+		return true // Always available for attempt
+	default:
+		return false
+	}
+}
+
+// GetConfiguredProviders returns a list of all providers with credentials configured
+func (c *AIConfig) GetConfiguredProviders() []string {
+	var providers []string
+	if c.HasProvider(AIProviderAnthropic) {
+		providers = append(providers, AIProviderAnthropic)
+	}
+	if c.HasProvider(AIProviderOpenAI) {
+		providers = append(providers, AIProviderOpenAI)
+	}
+	if c.HasProvider(AIProviderDeepSeek) {
+		providers = append(providers, AIProviderDeepSeek)
+	}
+	if c.HasProvider(AIProviderOllama) {
+		providers = append(providers, AIProviderOllama)
+	}
+	return providers
+}
+
+// GetAPIKeyForProvider returns the API key for the specified provider
+func (c *AIConfig) GetAPIKeyForProvider(provider string) string {
+	switch provider {
+	case AIProviderAnthropic:
+		if c.AnthropicAPIKey != "" {
+			return c.AnthropicAPIKey
+		}
+		// Fall back to legacy API key if provider matches
+		if c.Provider == AIProviderAnthropic {
+			return c.APIKey
+		}
+	case AIProviderOpenAI:
+		if c.OpenAIAPIKey != "" {
+			return c.OpenAIAPIKey
+		}
+		if c.Provider == AIProviderOpenAI {
+			return c.APIKey
+		}
+	case AIProviderDeepSeek:
+		if c.DeepSeekAPIKey != "" {
+			return c.DeepSeekAPIKey
+		}
+		if c.Provider == AIProviderDeepSeek {
+			return c.APIKey
+		}
+	}
+	return ""
+}
+
+// GetBaseURLForProvider returns the base URL for the specified provider
+func (c *AIConfig) GetBaseURLForProvider(provider string) string {
+	switch provider {
+	case AIProviderOllama:
+		if c.OllamaBaseURL != "" {
+			return c.OllamaBaseURL
+		}
+		// Fall back to legacy BaseURL if provider matches
+		if c.Provider == AIProviderOllama && c.BaseURL != "" {
+			return c.BaseURL
+		}
+		return DefaultOllamaBaseURL
+	case AIProviderOpenAI:
+		if c.OpenAIBaseURL != "" {
+			return c.OpenAIBaseURL
+		}
+		return "" // Uses default OpenAI URL
+	case AIProviderDeepSeek:
+		return DefaultDeepSeekBaseURL
+	}
+	return ""
+}
+
 // IsUsingOAuth returns true if OAuth authentication is configured for Anthropic
 func (c *AIConfig) IsUsingOAuth() bool {
-	return c.Provider == AIProviderAnthropic && c.AuthMethod == AuthMethodOAuth && c.OAuthAccessToken != ""
+	return c.AuthMethod == AuthMethodOAuth && c.OAuthAccessToken != ""
+}
+
+// ParseModelString parses a model string in "provider:model-name" format
+// Returns the provider and model name. If no provider prefix, attempts to detect.
+func ParseModelString(model string) (provider, modelName string) {
+	// Check for explicit provider prefix
+	for _, p := range []string{AIProviderAnthropic, AIProviderOpenAI, AIProviderDeepSeek, AIProviderOllama} {
+		prefix := p + ":"
+		if len(model) > len(prefix) && model[:len(prefix)] == prefix {
+			return p, model[len(prefix):]
+		}
+	}
+
+	// No prefix - try to detect from model name patterns
+	switch {
+	case len(model) >= 6 && model[:6] == "claude":
+		return AIProviderAnthropic, model
+	case len(model) >= 3 && (model[:3] == "gpt" || model[:2] == "o1" || model[:2] == "o3" || model[:2] == "o4"):
+		return AIProviderOpenAI, model
+	case len(model) >= 8 && model[:8] == "deepseek":
+		return AIProviderDeepSeek, model
+	default:
+		// Assume Ollama for unrecognized models (local models have varied names)
+		return AIProviderOllama, model
+	}
+}
+
+// FormatModelString creates a "provider:model-name" format string
+func FormatModelString(provider, modelName string) string {
+	return provider + ":" + modelName
 }
 
 // GetBaseURL returns the base URL, using defaults where appropriate
+// DEPRECATED: Use GetBaseURLForProvider instead
 func (c *AIConfig) GetBaseURL() string {
 	if c.BaseURL != "" {
 		return c.BaseURL
@@ -143,6 +295,24 @@ func (c *AIConfig) GetModel() string {
 	}
 }
 
+// GetChatModel returns the model for interactive chat conversations
+// Falls back to the main Model if ChatModel is not set
+func (c *AIConfig) GetChatModel() string {
+	if c.ChatModel != "" {
+		return c.ChatModel
+	}
+	return c.GetModel()
+}
+
+// GetPatrolModel returns the model for background patrol analysis
+// Falls back to the main Model if PatrolModel is not set
+func (c *AIConfig) GetPatrolModel() string {
+	if c.PatrolModel != "" {
+		return c.PatrolModel
+	}
+	return c.GetModel()
+}
+
 // ClearOAuthTokens clears OAuth tokens (used when switching back to API key auth)
 func (c *AIConfig) ClearOAuthTokens() {
 	c.OAuthAccessToken = ""