agentgui 1.0.274 → 1.0.275

package/lib/speech.js

@@ -1,152 +1,152 @@
-import { createRequire } from 'module';
-import fs from 'fs';
-import path from 'path';
-import os from 'os';
-import http from 'http';
-import { fileURLToPath } from 'url';
-
-const require = createRequire(import.meta.url);
-const __dirname = path.dirname(fileURLToPath(import.meta.url));
-const ROOT = path.dirname(__dirname);
-
-const serverSTT = require('webtalk/server-stt');
-const serverTTS = require('webtalk/server-tts');
-
-const EXTRA_VOICE_DIRS = [path.join(ROOT, 'voices')];
-
-const POCKET_TTS_VOICES = [
-  { id: 'default', name: 'Default', gender: 'female', accent: 'French' },
-  { id: 'alba', name: 'Alba', gender: 'female', accent: 'French' },
-  { id: 'marius', name: 'Marius', gender: 'male', accent: 'French' },
-  { id: 'javert', name: 'Javert', gender: 'male', accent: 'French' },
-  { id: 'jean', name: 'Jean', gender: 'male', accent: 'French' },
-  { id: 'fantine', name: 'Fantine', gender: 'female', accent: 'French' },
-  { id: 'cosette', name: 'Cosette', gender: 'female', accent: 'French' },
-  { id: 'eponine', name: 'Eponine', gender: 'female', accent: 'French' },
-  { id: 'azelma', name: 'Azelma', gender: 'female', accent: 'French' },
-];
-
-const PREDEFINED_IDS = new Set(POCKET_TTS_VOICES.filter(v => v.id !== 'default').map(v => v.id));
-const POCKET_PORT = 8787;
-
-function safeGetVoices(extraDirs) {
-  if (typeof serverTTS.getVoices === 'function') {
-    return serverTTS.getVoices(extraDirs || []);
-  }
-  return [];
-}
-
-const needsPatch = !safeGetVoices(EXTRA_VOICE_DIRS).some(v => v.id === 'alba' && !v.isCustom);
-
-function synthesizeDirect(text, voiceId) {
-  const voicePath = serverTTS.findVoiceFile(voiceId, EXTRA_VOICE_DIRS);
-  const isPredefined = voiceId && PREDEFINED_IDS.has(voiceId);
-  const boundary = '----PocketTTS' + Date.now();
-  const parts = [];
-  parts.push(`--${boundary}\r\nContent-Disposition: form-data; name="text"\r\n\r\n${text}\r\n`);
-  if (voicePath) {
-    const data = fs.readFileSync(voicePath);
-    const name = path.basename(voicePath);
-    parts.push(`--${boundary}\r\nContent-Disposition: form-data; name="voice_wav"; filename="${name}"\r\nContent-Type: audio/wav\r\n\r\n`);
-    parts.push(data);
-    parts.push('\r\n');
-  } else if (isPredefined) {
-    parts.push(`--${boundary}\r\nContent-Disposition: form-data; name="voice_url"\r\n\r\n${voiceId}\r\n`);
-  }
-  parts.push(`--${boundary}--\r\n`);
-  const body = Buffer.concat(parts.map(p => Buffer.isBuffer(p) ? p : Buffer.from(p)));
-  return new Promise((resolve, reject) => {
-    const req = http.request({
-      hostname: '127.0.0.1', port: POCKET_PORT, path: '/tts', method: 'POST',
-      headers: { 'Content-Type': `multipart/form-data; boundary=${boundary}`, 'Content-Length': body.length },
-      timeout: 60000,
-    }, res => {
-      if (res.statusCode !== 200) {
-        let e = '';
-        res.on('data', d => e += d);
-        res.on('end', () => reject(new Error(`pocket-tts HTTP ${res.statusCode}: ${e}`)));
-        return;
-      }
-      const chunks = [];
-      res.on('data', d => chunks.push(d));
-      res.on('end', () => resolve(Buffer.concat(chunks)));
-    });
-    req.on('error', reject);
-    req.on('timeout', () => { req.destroy(); reject(new Error('pocket-tts timeout')); });
-    req.write(body);
-    req.end();
-  });
-}
-
-function transcribe(audioBuffer) {
-  return serverSTT.transcribe(audioBuffer);
-}
-
-function getSTT() {
-  return serverSTT.getSTT();
-}
-
-function synthesize(text, voiceId) {
-  if (needsPatch && voiceId && PREDEFINED_IDS.has(voiceId)) {
-    return synthesizeDirect(text, voiceId);
-  }
-  return serverTTS.synthesize(text, voiceId, EXTRA_VOICE_DIRS);
-}
-
-function synthesizeStream(text, voiceId) {
-  if (needsPatch && voiceId && PREDEFINED_IDS.has(voiceId)) {
-    return (async function* () {
-      yield await synthesizeDirect(text, voiceId);
-    })();
-  }
-  return serverTTS.synthesizeStream(text, voiceId, EXTRA_VOICE_DIRS);
-}
-
-function getVoices() {
-  const upstream = safeGetVoices(EXTRA_VOICE_DIRS);
-  const custom = upstream.filter(v => v.isCustom);
-  return [...POCKET_TTS_VOICES, ...custom];
-}
-
-function getStatus() {
-  const sttStatus = serverSTT.getStatus();
-  const ttsStatus = serverTTS.getStatus();
-  return {
-    sttReady: sttStatus.ready,
-    ttsReady: ttsStatus.ready,
-    sttLoading: sttStatus.loading,
-    ttsLoading: false,
-    sttError: sttStatus.error,
-    ttsError: ttsStatus.ready ? null : (ttsStatus.lastError || 'pocket-tts not running'),
-    pocketTts: ttsStatus,
-  };
-}
-
-function preloadTTS() {
-  if (typeof serverTTS.findVoiceFile !== 'function' || typeof serverTTS.start !== 'function') {
-    console.log('[TTS] pocket-tts functions not available');
-    return;
-  }
-  const defaultVoice = serverTTS.findVoiceFile('custom_cleetus', EXTRA_VOICE_DIRS) || '/config/voices/cleetus.wav';
-  const voicePath = fs.existsSync(defaultVoice) ? defaultVoice : null;
-  serverTTS.start(voicePath, {}).then(ok => {
-    if (ok) console.log('[TTS] pocket-tts sidecar started');
-    else console.log('[TTS] pocket-tts failed to start');
-  }).catch(err => {
-    console.error('[TTS] pocket-tts start error:', err.message);
-  });
-}
-
-function ttsCacheKey(text, voiceId) {
-  return typeof serverTTS.ttsCacheKey === 'function' ? serverTTS.ttsCacheKey(text, voiceId) : null;
-}
-
-function ttsCacheGet(key) {
-  return typeof serverTTS.ttsCacheGet === 'function' ? serverTTS.ttsCacheGet(key) : null;
-}
-
-function splitSentences(text) {
-  return typeof serverTTS.splitSentences === 'function' ? serverTTS.splitSentences(text) : [text];
-}
-
-export { transcribe, synthesize, synthesizeStream, getSTT, getStatus, getVoices, preloadTTS, ttsCacheKey, ttsCacheGet, splitSentences };
+import { createRequire } from 'module';
+import fs from 'fs';
+import path from 'path';
+import os from 'os';
+import http from 'http';
+import { fileURLToPath } from 'url';
+
+const require = createRequire(import.meta.url);
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const ROOT = path.dirname(__dirname);
+
+const serverSTT = require('webtalk/server-stt');
+const serverTTS = require('webtalk/server-tts');
+
+const EXTRA_VOICE_DIRS = [path.join(ROOT, 'voices')];
+
+const POCKET_TTS_VOICES = [
+  { id: 'default', name: 'Default', gender: 'female', accent: 'French' },
+  { id: 'alba', name: 'Alba', gender: 'female', accent: 'French' },
+  { id: 'marius', name: 'Marius', gender: 'male', accent: 'French' },
+  { id: 'javert', name: 'Javert', gender: 'male', accent: 'French' },
+  { id: 'jean', name: 'Jean', gender: 'male', accent: 'French' },
+  { id: 'fantine', name: 'Fantine', gender: 'female', accent: 'French' },
+  { id: 'cosette', name: 'Cosette', gender: 'female', accent: 'French' },
+  { id: 'eponine', name: 'Eponine', gender: 'female', accent: 'French' },
+  { id: 'azelma', name: 'Azelma', gender: 'female', accent: 'French' },
+];
+
+const PREDEFINED_IDS = new Set(POCKET_TTS_VOICES.filter(v => v.id !== 'default').map(v => v.id));
+const POCKET_PORT = 8787;
+
+function safeGetVoices(extraDirs) {
+  if (typeof serverTTS.getVoices === 'function') {
+    return serverTTS.getVoices(extraDirs || []);
+  }
+  return [];
+}
+
+const needsPatch = !safeGetVoices(EXTRA_VOICE_DIRS).some(v => v.id === 'alba' && !v.isCustom);
+
+function synthesizeDirect(text, voiceId) {
+  const voicePath = serverTTS.findVoiceFile(voiceId, EXTRA_VOICE_DIRS);
+  const isPredefined = voiceId && PREDEFINED_IDS.has(voiceId);
+  const boundary = '----PocketTTS' + Date.now();
+  const parts = [];
+  parts.push(`--${boundary}\r\nContent-Disposition: form-data; name="text"\r\n\r\n${text}\r\n`);
+  if (voicePath) {
+    const data = fs.readFileSync(voicePath);
+    const name = path.basename(voicePath);
+    parts.push(`--${boundary}\r\nContent-Disposition: form-data; name="voice_wav"; filename="${name}"\r\nContent-Type: audio/wav\r\n\r\n`);
+    parts.push(data);
+    parts.push('\r\n');
+  } else if (isPredefined) {
+    parts.push(`--${boundary}\r\nContent-Disposition: form-data; name="voice_url"\r\n\r\n${voiceId}\r\n`);
+  }
+  parts.push(`--${boundary}--\r\n`);
+  const body = Buffer.concat(parts.map(p => Buffer.isBuffer(p) ? p : Buffer.from(p)));
+  return new Promise((resolve, reject) => {
+    const req = http.request({
+      hostname: '127.0.0.1', port: POCKET_PORT, path: '/tts', method: 'POST',
+      headers: { 'Content-Type': `multipart/form-data; boundary=${boundary}`, 'Content-Length': body.length },
+      timeout: 60000,
+    }, res => {
+      if (res.statusCode !== 200) {
+        let e = '';
+        res.on('data', d => e += d);
+        res.on('end', () => reject(new Error(`pocket-tts HTTP ${res.statusCode}: ${e}`)));
+        return;
+      }
+      const chunks = [];
+      res.on('data', d => chunks.push(d));
+      res.on('end', () => resolve(Buffer.concat(chunks)));
+    });
+    req.on('error', reject);
+    req.on('timeout', () => { req.destroy(); reject(new Error('pocket-tts timeout')); });
+    req.write(body);
+    req.end();
+  });
+}
+
+function transcribe(audioBuffer) {
+  return serverSTT.transcribe(audioBuffer);
+}
+
+function getSTT() {
+  return serverSTT.getSTT();
+}
+
+function synthesize(text, voiceId) {
+  if (needsPatch && voiceId && PREDEFINED_IDS.has(voiceId)) {
+    return synthesizeDirect(text, voiceId);
+  }
+  return serverTTS.synthesize(text, voiceId, EXTRA_VOICE_DIRS);
+}
+
+function synthesizeStream(text, voiceId) {
+  if (needsPatch && voiceId && PREDEFINED_IDS.has(voiceId)) {
+    return (async function* () {
+      yield await synthesizeDirect(text, voiceId);
+    })();
+  }
+  return serverTTS.synthesizeStream(text, voiceId, EXTRA_VOICE_DIRS);
+}
+
+function getVoices() {
+  const upstream = safeGetVoices(EXTRA_VOICE_DIRS);
+  const custom = upstream.filter(v => v.isCustom);
+  return [...POCKET_TTS_VOICES, ...custom];
+}
+
+function getStatus() {
+  const sttStatus = serverSTT.getStatus();
+  const ttsStatus = serverTTS.getStatus();
+  return {
+    sttReady: sttStatus.ready,
+    ttsReady: ttsStatus.ready,
+    sttLoading: sttStatus.loading,
+    ttsLoading: false,
+    sttError: sttStatus.error,
+    ttsError: ttsStatus.ready ? null : (ttsStatus.lastError || 'pocket-tts not running'),
+    pocketTts: ttsStatus,
+  };
+}
+
+function preloadTTS() {
+  if (typeof serverTTS.findVoiceFile !== 'function' || typeof serverTTS.start !== 'function') {
+    console.log('[TTS] pocket-tts functions not available');
+    return;
+  }
+  const defaultVoice = serverTTS.findVoiceFile('custom_cleetus', EXTRA_VOICE_DIRS) || '/config/voices/cleetus.wav';
+  const voicePath = fs.existsSync(defaultVoice) ? defaultVoice : null;
+  serverTTS.start(voicePath, {}).then(ok => {
+    if (ok) console.log('[TTS] pocket-tts sidecar started');
+    else console.log('[TTS] pocket-tts failed to start');
+  }).catch(err => {
+    console.error('[TTS] pocket-tts start error:', err.message);
+  });
+}
+
+function ttsCacheKey(text, voiceId) {
+  return typeof serverTTS.ttsCacheKey === 'function' ? serverTTS.ttsCacheKey(text, voiceId) : null;
+}
+
+function ttsCacheGet(key) {
+  return typeof serverTTS.ttsCacheGet === 'function' ? serverTTS.ttsCacheGet(key) : null;
+}
+
+function splitSentences(text) {
+  return typeof serverTTS.splitSentences === 'function' ? serverTTS.splitSentences(text) : [text];
+}
+
+export { transcribe, synthesize, synthesizeStream, getSTT, getStatus, getVoices, preloadTTS, ttsCacheKey, ttsCacheGet, splitSentences };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agentgui",
-  "version": "1.0.274",
+  "version": "1.0.275",
   "description": "Multi-agent ACP client with real-time communication",
   "type": "module",
   "main": "server.js",

package/readme.md

@@ -1,76 +1,76 @@
-# AgentGUI
-
-A multi-agent GUI for AI coding assistants. Connects to CLI-based agents (Claude Code, Gemini CLI, OpenCode, Goose, and others) and provides a web interface with real-time streaming output.
-
-## Quick Start
-
-```bash
-npx agentgui
-```
-
-Or install and run manually:
-
-```bash
-git clone https://github.com/AnEntrypoint/agentgui.git
-cd agentgui
-npm install
-npm run dev
-```
-
-Open `http://localhost:3000` in your browser.
-
-## What It Does
-
-- Auto-discovers AI coding agents installed on your system (Claude Code, Gemini CLI, OpenCode, Goose, Codex, Kiro, etc.)
-- Runs agents with streaming JSON output and displays results in real-time via WebSocket
-- Manages conversations with SQLite persistence
-- Supports concurrent agent sessions
-- Provides file browsing and upload for agent working directories
-- Includes speech-to-text and text-to-speech
-
-## Architecture
-
-- `server.js` - HTTP server, REST API, WebSocket endpoint, static file serving
-- `database.js` - SQLite database (WAL mode) at `~/.gmgui/data.db`
-- `lib/claude-runner.js` - Agent runner framework, spawns CLI processes and parses streaming output
-- `lib/speech.js` - Speech processing via Hugging Face transformers
-- `static/` - Browser client with streaming renderer, WebSocket manager, and HTML templates
-- `bin/gmgui.cjs` - CLI entry point for `npx agentgui`
-
-## Text-to-Speech on Windows
-
-On Windows, AgentGUI automatically sets up pocket-tts (text-to-speech) on your first TTS request. No manual setup required.
-
-### What Happens
-1. Server detects Python 3.9+ installation
-2. Creates virtual environment at `~/.gmgui/pocket-venv`
-3. Installs pocket-tts via pip
-4. All subsequent TTS requests use cached installation
-
-### Requirements
-- Python 3.9+ (check with `python --version`)
-- ~200 MB free disk space
-- Internet connection for first setup
-
-### Troubleshooting
-- **Python not found**: Download from https://www.python.org and ensure "Add Python to PATH" is checked
-- **Setup fails**: Check that you have write access to your home directory (~/.gmgui/)
-- **Manual cleanup**: Delete `%USERPROFILE%\.gmgui\pocket-venv` and try again
-
-For manual setup or detailed troubleshooting, see the setup instructions in the code or check `/api/speech-status` endpoint for error details.
-
-## Configuration
-
-| Variable | Default | Description |
-|----------|---------|-------------|
-| `PORT` | 3000 | Server port |
-| `BASE_URL` | /gm | URL prefix |
-| `HOT_RELOAD` | true | Watch mode for development |
-
-## License
-
-MIT
-
-## Repository
-
-https://github.com/AnEntrypoint/agentgui
+# AgentGUI
+
+A multi-agent GUI for AI coding assistants. Connects to CLI-based agents (Claude Code, Gemini CLI, OpenCode, Goose, and others) and provides a web interface with real-time streaming output.
+
+## Quick Start
+
+```bash
+npx agentgui
+```
+
+Or install and run manually:
+
+```bash
+git clone https://github.com/AnEntrypoint/agentgui.git
+cd agentgui
+npm install
+npm run dev
+```
+
+Open `http://localhost:3000` in your browser.
+
+## What It Does
+
+- Auto-discovers AI coding agents installed on your system (Claude Code, Gemini CLI, OpenCode, Goose, Codex, Kiro, etc.)
+- Runs agents with streaming JSON output and displays results in real-time via WebSocket
+- Manages conversations with SQLite persistence
+- Supports concurrent agent sessions
+- Provides file browsing and upload for agent working directories
+- Includes speech-to-text and text-to-speech
+
+## Architecture
+
+- `server.js` - HTTP server, REST API, WebSocket endpoint, static file serving
+- `database.js` - SQLite database (WAL mode) at `~/.gmgui/data.db`
+- `lib/claude-runner.js` - Agent runner framework, spawns CLI processes and parses streaming output
+- `lib/speech.js` - Speech processing via Hugging Face transformers
+- `static/` - Browser client with streaming renderer, WebSocket manager, and HTML templates
+- `bin/gmgui.cjs` - CLI entry point for `npx agentgui`
+
+## Text-to-Speech on Windows
+
+On Windows, AgentGUI automatically sets up pocket-tts (text-to-speech) on your first TTS request. No manual setup required.
+
+### What Happens
+1. Server detects Python 3.9+ installation
+2. Creates virtual environment at `~/.gmgui/pocket-venv`
+3. Installs pocket-tts via pip
+4. All subsequent TTS requests use cached installation
+
+### Requirements
+- Python 3.9+ (check with `python --version`)
+- ~200 MB free disk space
+- Internet connection for first setup
+
+### Troubleshooting
+- **Python not found**: Download from https://www.python.org and ensure "Add Python to PATH" is checked
+- **Setup fails**: Check that you have write access to your home directory (~/.gmgui/)
+- **Manual cleanup**: Delete `%USERPROFILE%\.gmgui\pocket-venv` and try again
+
+For manual setup or detailed troubleshooting, see the setup instructions in the code or check `/api/speech-status` endpoint for error details.
+
+## Configuration
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `PORT` | 3000 | Server port |
+| `BASE_URL` | /gm | URL prefix |
+| `HOT_RELOAD` | true | Watch mode for development |
+
+## License
+
+MIT
+
+## Repository
+
+https://github.com/AnEntrypoint/agentgui