Feat/v2 spotlight

README.md

@@ -1,6 +1,8 @@
 # RouteBox
 
-macOS menu bar app — LLM API proxy with intelligent routing, real-time monitoring, and cost tracking.
+macOS menu bar app — LLM API proxy with intelligent routing, real-time monitoring, cost tracking, built-in chat, and web search.
+
+**v2.3.0** · [Download DMG](https://github.com/createpjf/RouteBox/releases/latest) · MIT License
 
 ## What it does
 
@@ -9,9 +11,12 @@ RouteBox runs a local OpenAI-compatible proxy (`http://localhost:3001/v1`). You
 1. **Routes** requests to the best provider based on your rules (cheapest, fastest, or smartest)
 2. **Tracks** tokens, cost, latency, and savings in real-time
 3. **Manages** multiple provider API keys securely in macOS Keychain
+4. **Chat** directly with any model via the built-in Chat window or Spotlight
+5. **Search** the web in real-time with Brave Search integration
 
 ```
 Your App  →  RouteBox (localhost:3001)  →  OpenAI / Anthropic / Google / DeepSeek / MiniMax / Kimi / FLock.io
+                                           ↑ Brave Search results injected as context
 ```
 
 ## Supported Providers
@@ -26,7 +31,66 @@ Your App  →  RouteBox (localhost:3001)  →  OpenAI / Anthropic / Google / Dee
 | [Kimi](https://kimi.ai) | Kimi K2.5, Kimi K2, Moonshot | [platform.moonshot.ai](https://platform.moonshot.ai/) |
 | [FLock.io](https://flock.io) | Qwen3-235B, Qwen3-30B, DeepSeek-V3.2, Kimi K2.5 | [platform.flock.io](https://platform.flock.io) |
 
-> **Tip:** [FLock API Platform](https://platform.flock.io) provides access to open-source models (Qwen3, DeepSeek, Kimi)— a good option for cost-effective routing.
+**Local models**: Ollama and LM Studio are auto-discovered on your network — no API key needed.
+
+> **Tip:** [FLock API Platform](https://platform.flock.io) provides access to open-source models (Qwen3, DeepSeek, Kimi) — a good option for cost-effective routing.
+
+## Features
+
+### Three Windows
+
+| Window | Shortcut | Description |
+|--------|----------|-------------|
+| **Panel** | `⌘⇧R` | Menu bar dashboard — routing, analytics, logs, settings |
+| **Spotlight** | `⌘⇧Space` | Quick floating window — ask a question, get an instant answer |
+| **Chat** | Via panel | Full chat interface with conversation history and sidebar |
+
+### Dashboard Tabs
+
+| Tab | What it shows |
+|-----|---------------|
+| **Dashboard** | Requests, tokens, cost, savings, traffic sparkline, provider status |
+| **Routing** | Strategy selector, model preferences (pin/exclude), content-aware rules |
+| **Logs** | Full request history with model, provider, latency, cost per request |
+| **Analytics** | Charts for cost trends, provider latency, model usage breakdown |
+| **My Usage** | Today/month usage, budget tracking, weekly trends, model breakdown |
+
+### Web Search (Brave Search)
+
+RouteBox can inject real-time web search results into any LLM conversation:
+
+1. Go to **Settings → Web Search** and add your [Brave Search API key](https://brave.com/search/api) (free tier available)
+2. Toggle the 🌐 button in Chat or Spotlight to enable search for a message
+3. RouteBox searches the web, injects results as context, and the model responds with up-to-date information and source citations
+
+### Intelligent Routing
+
+| Strategy | Behavior |
+|----------|----------|
+| Smart Auto | AI picks the best route per request |
+| Cost First | Always pick the cheapest provider |
+| Speed First | Always pick the lowest latency provider |
+| Quality First | Always pick the best available model |
+
+### Routing Rules
+
+| Rule Type | Triggers when... | Example use |
+|-----------|-------------------|-------------|
+| **Alias** | Model name matches your virtual name | `route-code` → `deepseek-coder` |
+| **Code** | Request contains ≥3 code markers | Auto-route code tasks to DeepSeek |
+| **Long** | Message ≥8,000 characters | Auto-route long context to Gemini |
+| **General** | Catch-all fallback | Default model for everything else |
+
+### Model Preferences
+
+Pin a model to a specific provider, or exclude a provider for a model:
+
+- **Pin**: `gpt-4o` → always use OpenAI (never fall back)
+- **Exclude**: `gpt-4o` → never use provider X
+
+### Thinking Model Support
+
+Models that output `<think>` blocks (DeepSeek-R1, Qwen3, etc.) are automatically handled — thinking content is hidden behind a collapsible "💭 Thinking..." section.
 
 ## Setup
 
@@ -78,45 +142,11 @@ client = OpenAI(
 )
 ```
 
-## App Tabs
-
-| Tab | What it shows |
-|-----|---------------|
-| **Dashboard** | Requests, tokens, cost, savings, traffic sparkline, provider status |
-| **Routing** | Strategy selector, model preferences (pin/exclude), content-aware rules |
-| **Logs** | Full request history with model, provider, latency, cost per request |
-| **Analytics** | Charts for cost trends, provider latency, model usage breakdown |
-
-## Routing
-
-### Strategy
-
-Pick one in the Routing tab:
-
-| Strategy | Behavior |
-|----------|----------|
-| Smart Auto | AI picks the best route per request |
-| Cost First | Always pick the cheapest provider |
-| Speed First | Always pick the lowest latency provider |
-| Quality First | Always pick the best available model |
-
-### Rules
-
-Create rules to route specific request types:
+### 4. Enable Web Search (Optional)
 
-| Rule Type | Triggers when... | Example use |
-|-----------|-------------------|-------------|
-| **Alias** | Model name matches your virtual name | `route-code` → `deepseek-coder` |
-| **Code** | Request contains ≥3 code markers | Auto-route code tasks to DeepSeek |
-| **Long** | Message ≥8,000 characters | Auto-route long context to Gemini |
-| **General** | Catch-all fallback | Default model for everything else |
+Go to **Settings → Web Search** → Paste your Brave Search API key → Save.
 
-### Model Preferences
-
-Pin a model to a specific provider, or exclude a provider for a model:
-
-- **Pin**: `gpt-4o` → always use OpenAI (never fall back)
-- **Exclude**: `gpt-4o` → never use provider X
+Then toggle 🌐 in Chat or Spotlight before sending a message to include web results.
 
 ## Build DMG
 
@@ -146,6 +176,7 @@ docker build -t routebox-gateway .
 docker run -p 3001:3001 \
   -e OPENAI_API_KEY=sk-... \
   -e ANTHROPIC_API_KEY=sk-ant-... \
+  -e BRAVE_API_KEY=BSA-... \
   -v routebox-data:/data \
   routebox-gateway
 ```
@@ -158,6 +189,7 @@ See [`apps/gateway/.env.example`](apps/gateway/.env.example) for all environment
 |---------|----------|-------|
 | Provider API Keys | Settings → Providers | Stored in macOS Keychain |
 | Monthly Budget | Settings → Budget | Alerts at 80% and 100% |
+| Web Search | Settings → Web Search | Brave Search API key for real-time search |
 | Gateway URL | Settings → Connection | Default `http://localhost:3001`, customizable |
 | Auth Token | Settings → Authentication | Auto-generated, stored in Keychain |
 | Auto-start Gateway | Settings → Gateway | On/off toggle |
@@ -168,13 +200,29 @@ See [`apps/gateway/.env.example`](apps/gateway/.env.example) for all environment
 | Shortcut | Action |
 |----------|--------|
 | `⌘⇧R` | Toggle panel (global) |
+| `⌘⇧Space` | Toggle Spotlight (global) |
 | `⌘C` | Copy API key |
 | `⌘P` | Pause/resume traffic |
+| `⌘⏎` | Send message (Spotlight) |
+| `Esc` | Close Spotlight |
+
+## Architecture
+
+```
+RouteBox/
+├── apps/
+│   ├── desktop/          Tauri v2 + React 19 (panel, spotlight, chat windows)
+│   │   └── src-tauri/    Rust backend (system tray, global shortcuts, keychain)
+│   └── gateway/          Bun + Hono (proxy, routing, analytics, search)
+├── package.json          pnpm monorepo root
+└── README.md
+```
 
 ## Tech Stack
 
 - **Desktop**: Tauri v2 (Rust) + React 19 + TypeScript + Tailwind CSS v4
 - **Gateway**: Bun + Hono + bun:sqlite
+- **Search**: Brave Search API
 - **Design**: SF Pro, frosted glass (macOS native)
 
 ## License

apps/desktop/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@routebox/desktop",
   "private": true,
-  "version": "1.2.0",
+  "version": "2.3.0",
   "type": "module",
   "scripts": {
     "dev": "vite",
@@ -23,7 +23,10 @@
     "lucide-react": "^0.460.0",
     "react": "^19.1.0",
     "react-dom": "^19.1.0",
-    "recharts": "^2.15.0"
+    "react-markdown": "^9.1.0",
+    "recharts": "^2.15.0",
+    "rehype-highlight": "^7.0.2",
+    "remark-gfm": "^4.0.1"
   },
   "devDependencies": {
     "@tailwindcss/postcss": "^4.0.0",

apps/desktop/src-tauri/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "routebox-desktop"
-version = "1.2.0"
+version = "2.3.0"
 description = "RouteBox macOS Menu Bar App"
 authors = ["RouteBox"]
 edition = "2021"

apps/desktop/src-tauri/capabilities/default.json

@@ -10,6 +10,7 @@
     "global-shortcut:default",
     "notification:default",
     "store:default",
-    "process:default"
+    "process:default",
+    "updater:default"
   ]
 }

apps/desktop/src-tauri/src/commands.rs

@@ -1,5 +1,5 @@
 use crate::keychain;
-use tauri::Manager;
+use tauri::{Emitter, Manager};
 use tauri_plugin_positioner::{Position, WindowExt};
 use arboard::Clipboard;
 use std::sync::Mutex;
@@ -70,6 +70,62 @@ pub fn toggle_panel_internal(app: &tauri::AppHandle) -> Result<(), String> {
     Ok(())
 }
 
+// ── Spotlight toggle ────────────────────────────────────────────────────────
+
+#[tauri::command]
+pub async fn toggle_spotlight(app: tauri::AppHandle) -> Result<(), String> {
+    toggle_spotlight_internal(&app)
+}
+
+pub fn toggle_spotlight_internal(app: &tauri::AppHandle) -> Result<(), String> {
+    if let Some(window) = app.get_webview_window("spotlight") {
+        if window.is_visible().unwrap_or(false) {
+            window.hide().map_err(|e| e.to_string())?;
+        } else {
+            window.center().map_err(|e| e.to_string())?;
+            window.show().map_err(|e| e.to_string())?;
+            window.set_focus().map_err(|e| e.to_string())?;
+        }
+    }
+    Ok(())
+}
+
+// ── Chat window ─────────────────────────────────────────────────────────────
+
+#[tauri::command]
+pub async fn open_chat(app: tauri::AppHandle) -> Result<(), String> {
+    if let Some(window) = app.get_webview_window("chat") {
+        window.show().map_err(|e| e.to_string())?;
+        window.set_focus().map_err(|e| e.to_string())?;
+    }
+    Ok(())
+}
+
+// ── Clipboard read ──────────────────────────────────────────────────────────
+
+#[tauri::command]
+pub async fn read_clipboard() -> Result<String, String> {
+    let mut clipboard = Clipboard::new().map_err(|e| e.to_string())?;
+    clipboard.get_text().map_err(|e| e.to_string())
+}
+
+// ── Clipboard action (read + emit event + show spotlight) ───────────────────
+
+#[tauri::command]
+pub async fn clipboard_action(app: tauri::AppHandle, action: String) -> Result<(), String> {
+    clipboard_action_sync(&app, &action)
+}
+
+/// Synchronous version for use from shortcut handlers (no async runtime needed)
+pub fn clipboard_action_sync(app: &tauri::AppHandle, action: &str) -> Result<(), String> {
+    let mut clipboard = Clipboard::new().map_err(|e| e.to_string())?;
+    let text = clipboard.get_text().unwrap_or_default();
+    app.emit("spotlight-action", serde_json::json!({ "action": action, "text": text }))
+        .map_err(|e: tauri::Error| e.to_string())?;
+    toggle_spotlight_internal(app)?;
+    Ok(())
+}
+
 // ── Gateway process management ──────────────────────────────────────────────
 
 #[tauri::command]

apps/desktop/src-tauri/src/lib.rs

@@ -29,8 +29,6 @@ pub fn run() {
                 window.on_window_event(move |event| {
                     if let tauri::WindowEvent::Focused(false) = event {
                         let win = w.clone();
-                        // Small delay: if focus returns within 200ms (e.g. DevTools, HMR),
-                        // don't hide. This prevents the panel flickering in dev mode.
                         std::thread::spawn(move || {
                             std::thread::sleep(std::time::Duration::from_millis(200));
                             if !win.is_focused().unwrap_or(true) {
@@ -41,27 +39,70 @@ pub fn run() {
                 });
             }
 
-            // Register global hotkey: Cmd+Shift+R
+            // Hide spotlight when it loses focus
+            if let Some(window) = app.get_webview_window("spotlight") {
+                let w = window.clone();
+                window.on_window_event(move |event| {
+                    if let tauri::WindowEvent::Focused(false) = event {
+                        let win = w.clone();
+                        std::thread::spawn(move || {
+                            std::thread::sleep(std::time::Duration::from_millis(200));
+                            if !win.is_focused().unwrap_or(true) {
+                                let _ = win.hide();
+                            }
+                        });
+                    }
+                });
+            }
+
+            // Register global hotkeys
             #[cfg(desktop)]
             {
                 use tauri_plugin_global_shortcut::{
                     Code, GlobalShortcutExt, Modifiers, Shortcut, ShortcutState,
                 };
 
-                let shortcut =
+                let shortcut_panel =
                     Shortcut::new(Some(Modifiers::META | Modifiers::SHIFT), Code::KeyR);
+                let shortcut_spotlight =
+                    Shortcut::new(Some(Modifiers::META | Modifiers::SHIFT), Code::KeyX);
+                let shortcut_translate =
+                    Shortcut::new(Some(Modifiers::META | Modifiers::SHIFT), Code::KeyT);
+                let shortcut_summarize =
+                    Shortcut::new(Some(Modifiers::META | Modifiers::SHIFT), Code::KeyS);
+                let shortcut_explain =
+                    Shortcut::new(Some(Modifiers::META | Modifiers::SHIFT), Code::KeyE);
 
                 let handle = app.handle().clone();
+                let sc_panel = shortcut_panel.clone();
+                let sc_spotlight = shortcut_spotlight.clone();
+                let sc_translate = shortcut_translate.clone();
+                let sc_summarize = shortcut_summarize.clone();
+
                 app.handle().plugin(
                     tauri_plugin_global_shortcut::Builder::new()
-                        .with_handler(move |_app, _shortcut, event| {
+                        .with_handler(move |_app, shortcut, event| {
                             if event.state() == ShortcutState::Pressed {
-                                let _ = commands::toggle_panel_internal(&handle);
+                                if *shortcut == sc_panel {
+                                    let _ = commands::toggle_panel_internal(&handle);
+                                } else if *shortcut == sc_spotlight {
+                                    let _ = commands::toggle_spotlight_internal(&handle);
+                                } else if *shortcut == sc_translate {
+                                    let _ = commands::clipboard_action_sync(&handle, "translate");
+                                } else if *shortcut == sc_summarize {
+                                    let _ = commands::clipboard_action_sync(&handle, "summarize");
+                                } else {
+                                    let _ = commands::clipboard_action_sync(&handle, "explain");
+                                }
                             }
                         })
                         .build(),
                 )?;
-                app.global_shortcut().register(shortcut)?;
+                app.global_shortcut().register(shortcut_panel)?;
+                app.global_shortcut().register(shortcut_spotlight)?;
+                app.global_shortcut().register(shortcut_translate)?;
+                app.global_shortcut().register(shortcut_summarize)?;
+                app.global_shortcut().register(shortcut_explain)?;
             }
 
             Ok(())
@@ -73,6 +114,10 @@ pub fn run() {
             commands::copy_to_clipboard,
             commands::show_notification,
             commands::toggle_panel,
+            commands::toggle_spotlight,
+            commands::open_chat,
+            commands::read_clipboard,
+            commands::clipboard_action,
             commands::spawn_gateway,
             commands::stop_gateway,
             commands::is_gateway_running,