omni-notify-mcp 1.1.4 → 1.1.5

package/README.md

@@ -86,7 +86,7 @@ Priority routing for `notify`:
 ## Features
 
 ### Channels
-- **Desktop** — native `node-notifier` (macOS/Windows/Linux). Per-channel **system-sound toggle**.
+- **Desktop** — native `node-notifier` (macOS/Windows/Linux). Per-channel **system-sound toggle** and optional **text-to-speech** (natural neural voice via `msedge-tts`, no API key).
 - **Telegram** — bidirectional. The bot **replies in-thread** to user messages and acknowledges every inbound message so the user knows it landed.
 - **SMS** — Twilio.
 - **Email** — Gmail App Password (one click) or any SMTP. `ask` over email sends a reply link the user clicks to answer.

package/dist/channels/desktop.js

@@ -1,5 +1,8 @@
 import notifier from "node-notifier";
 import { spawn } from "child_process";
+import { tmpdir } from "os";
+import { join } from "path";
+const DEFAULT_TTS_VOICE = "en-US-AndrewMultilingualNeural";
 export async function sendDesktop(config, message) {
     if (!config.enabled)
         return;
@@ -12,7 +15,34 @@ export async function sendDesktop(config, message) {
             "Add-Type -AssemblyName System.Windows.Forms; [System.Media.SystemSounds]::Asterisk.Play(); Start-Sleep -Milliseconds 600",
         ], { windowsHide: true, stdio: "ignore" });
     }
+    if (config.tts) {
+        // Fire-and-forget: synthesize and play in the background so we don't
+        // block the MCP request on network + audio playback.
+        speak(message, config.ttsVoice ?? DEFAULT_TTS_VOICE).catch(() => { });
+    }
     await new Promise((resolve, reject) => {
         notifier.notify({ title: "Claude", message, sound: wantSound && process.platform !== "win32" }, (err) => err ? reject(err) : resolve());
     });
 }
+export async function speak(text, voice = DEFAULT_TTS_VOICE) {
+    // Lazy import so boot doesn't pay for msedge-tts when TTS is off.
+    const mod = await import("msedge-tts");
+    const { MsEdgeTTS, OUTPUT_FORMAT } = mod;
+    const tts = new MsEdgeTTS();
+    await tts.setMetadata(voice, OUTPUT_FORMAT.AUDIO_24KHZ_48KBITRATE_MONO_MP3);
+    const { mkdtempSync } = await import("fs");
+    const outDir = mkdtempSync(join(tmpdir(), "notify-tts-"));
+    const { audioFilePath } = await tts.toFile(outDir, text);
+    if (process.platform === "win32") {
+        spawn("powershell", [
+            "-NoProfile", "-Command",
+            `Add-Type -AssemblyName presentationCore; $p = New-Object System.Windows.Media.MediaPlayer; $p.Open([uri]'${audioFilePath.replace(/\\/g, "\\\\")}'); $p.Play(); Start-Sleep -Seconds 10`,
+        ], { windowsHide: true, stdio: "ignore" });
+    }
+    else if (process.platform === "darwin") {
+        spawn("afplay", [audioFilePath], { stdio: "ignore" });
+    }
+    else {
+        spawn("aplay", [audioFilePath], { stdio: "ignore" });
+    }
+}

package/dist/ui/server.js

@@ -13,6 +13,7 @@ import twilio from "twilio";
 import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
 import { StreamableHTTPServerTransport } from "@modelcontextprotocol/sdk/server/streamableHttp.js";
 import { z } from "zod";
+import { tmpdir } from "os";
 const PORT = process.env.PORT ? parseInt(process.env.PORT) : 3737;
 const REDIRECT_URI = `http://localhost:${PORT}/auth/google/callback`;
 const PUBLIC_DIR = join(fileURLToPath(new URL("../../ui/public", import.meta.url)));
@@ -168,6 +169,38 @@ app.post("/api/test/sound", (_req, res) => {
             res.json({ ok: true, message: "System sound triggered" });
     });
 });
+async function speakText(text, voice) {
+    const mod = await import("msedge-tts");
+    const { MsEdgeTTS, OUTPUT_FORMAT } = mod;
+    const tts = new MsEdgeTTS();
+    await tts.setMetadata(voice, OUTPUT_FORMAT.AUDIO_24KHZ_48KBITRATE_MONO_MP3);
+    const { mkdtempSync } = await import("fs");
+    const outDir = mkdtempSync(join(tmpdir(), "notify-tts-"));
+    const { audioFilePath } = await tts.toFile(outDir, text);
+    if (process.platform === "win32") {
+        spawn("powershell", [
+            "-NoProfile", "-Command",
+            `Add-Type -AssemblyName presentationCore; $p = New-Object System.Windows.Media.MediaPlayer; $p.Open([uri]'${audioFilePath.replace(/\\/g, "\\\\")}'); $p.Play(); Start-Sleep -Seconds 10`,
+        ], { windowsHide: true, stdio: "ignore" });
+    }
+    else if (process.platform === "darwin") {
+        spawn("afplay", [audioFilePath], { stdio: "ignore" });
+    }
+    else {
+        spawn("aplay", [audioFilePath], { stdio: "ignore" });
+    }
+}
+app.post("/api/test/tts", async (_req, res) => {
+    try {
+        const cfg = loadConfig();
+        const voice = cfg.desktop?.ttsVoice ?? "en-US-AndrewMultilingualNeural";
+        await speakText("Notification from Claude. This is a voice test.", voice);
+        res.json({ ok: true, message: `TTS played (${voice})` });
+    }
+    catch (err) {
+        res.status(500).json({ error: err instanceof Error ? err.message : String(err) });
+    }
+});
 app.post("/api/test/desktop", (_req, res) => {
     const time = new Date().toLocaleTimeString();
     const cfg = loadConfig();
@@ -568,6 +601,10 @@ async function sendNotification(message, priority, client) {
                 ], { windowsHide: true, stdio: "ignore" });
             }
             const soundOpt = wantSound && process.platform !== "win32";
+            if (cfg.desktop?.tts) {
+                const voice = cfg.desktop?.ttsVoice ?? "en-US-AndrewMultilingualNeural";
+                speakText(message, voice).catch((err) => log("→", "tts", `ERROR: ${err instanceof Error ? err.message : String(err)}`, client));
+            }
             await send("desktop", () => new Promise((res, rej) => notifier.notify({ title: "Claude Notify", message, sound: soundOpt }, (err) => err ? rej(err) : res())));
         }
         if (!desktopOnlyMode && cfg.telegram?.enabled && cfg.telegram.token && cfg.telegram.chatId) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "omni-notify-mcp",
-  "version": "1.1.4",
+  "version": "1.1.5",
   "description": "An MCP server that lets AI agents (Claude, Cursor, etc.) reach you on any channel — desktop, Telegram, SMS, email — with two-way ask/reply, real-time inbox push, Do Not Disturb, idle gating, multi-session routing, and a one-page web UI for setup. Zero config code; configure once, agents call notify/ask.",
   "main": "dist/index.js",
   "type": "module",
@@ -53,6 +53,7 @@
     "@modelcontextprotocol/sdk": "^1.0.0",
     "express": "^4.19.2",
     "googleapis": "^144.0.0",
+    "msedge-tts": "^2.0.5",
     "node-notifier": "^10.0.1",
     "nodemailer": "^6.9.9",
     "open": "^10.1.0",

package/ui/public/app.js

@@ -44,6 +44,7 @@ function populateForm() {
   // Desktop
   $("desktop-enabled").checked = !!config.desktop?.enabled;
   $("desktop-sound").checked = config.desktop?.sound !== false; // default on
+  $("desktop-tts").checked = !!config.desktop?.tts; // default off
 
   // Email / Gmail
   const email = config.email ?? {};
@@ -154,6 +155,7 @@ function saveDesktop() {
     desktop: {
       enabled: $("desktop-enabled").checked,
       sound: $("desktop-sound").checked,
+      tts: $("desktop-tts").checked,
     },
   });
 }
@@ -398,6 +400,17 @@ async function testSound() {
   }
 }
 
+async function testTts() {
+  try {
+    const res = await fetch("/api/test/tts", { method: "POST" });
+    const json = await res.json();
+    if (!res.ok) throw new Error(json.error);
+    toast(json.message, "ok");
+  } catch (e) {
+    toast("TTS test failed: " + e, "error");
+  }
+}
+
 // ── Test channels ─────────────────────────────────────────────────────────
 
 async function testChannel(channel) {

package/ui/public/index.html

@@ -55,6 +55,14 @@
             <span class="toggle-lbl">Play system sound</span>
             <button class="btn btn-sm btn-ghost" onclick="testSound()">Test sound</button>
           </div>
+          <div class="actions" style="margin-top:6px">
+            <label class="toggle-wrap">
+              <input type="checkbox" id="desktop-tts" onchange="saveDesktop()">
+              <span class="toggle"></span>
+            </label>
+            <span class="toggle-lbl">Speak notification</span>
+            <button class="btn btn-sm btn-ghost" onclick="testTts()">Test voice</button>
+          </div>
           <span id="os-hint" class="os-tag hidden"></span>
         </div>
       </div>