openvoiceui 1.0.0

package/auto-approve-devices.js

@@ -0,0 +1,111 @@
+#!/usr/bin/env node
+// auto-approve-devices.js — Approve all pending OpenClaw devices.
+// Runs on the host, execs into the openclaw container to move pending → paired.
+// Called by start.js after containers are healthy.
+//
+// Polls for up to 2 minutes (24 retries × 5s) because OpenVoiceUI doesn't
+// connect to OpenClaw until the user opens the browser — which happens AFTER
+// Pinokio shows the "Open" button. We need to wait for that.
+
+const { execSync } = require("child_process");
+
+const COMPOSE = "docker compose -f docker-compose.yml -f docker-compose.pinokio.yml";
+const MAX_ATTEMPTS = 24;  // 24 × 5s = 2 minutes
+const DELAY_MS = 5000;
+
+// MSYS_NO_PATHCONV=1 prevents Git Bash on Windows from converting /container/paths
+const EXEC_ENV = Object.assign({}, process.env, { MSYS_NO_PATHCONV: "1" });
+
+// Node one-liner that runs INSIDE the openclaw container
+const script = `
+const fs = require('fs');
+try {
+  const pendingPath = '/root/.openclaw/devices/pending.json';
+  const pairedPath = '/root/.openclaw/devices/paired.json';
+  let pending = {};
+  let paired = {};
+  try { pending = JSON.parse(fs.readFileSync(pendingPath, 'utf8')); } catch(e) {}
+  try { paired = JSON.parse(fs.readFileSync(pairedPath, 'utf8')); } catch(e) {}
+  const pairedCount = Object.keys(paired).length;
+  let count = 0;
+  for (const entry of Object.values(pending)) {
+    if (entry.deviceId && entry.publicKey) {
+      paired[entry.deviceId] = {
+        publicKey: entry.publicKey,
+        name: entry.name || 'auto-approved',
+        role: entry.role || 'operator',
+        roles: entry.roles || ['operator'],
+        scopes: entry.scopes || ['operator.read', 'operator.write'],
+        paired: true,
+        pairedAt: new Date().toISOString(),
+        autoApproved: true,
+      };
+      count++;
+    }
+  }
+  if (count > 0) {
+    fs.writeFileSync(pairedPath, JSON.stringify(paired, null, 2));
+    fs.writeFileSync(pendingPath, '{}');
+    console.log('APPROVED:' + count);
+  } else if (pairedCount > 0) {
+    console.log('ALREADY_PAIRED:' + pairedCount);
+  } else {
+    console.log('APPROVED:0');
+  }
+} catch(e) {
+  console.log('ERROR:' + e.message);
+}
+`.replace(/\n/g, " ").trim();
+
+function sleep(ms) {
+  return new Promise(resolve => setTimeout(resolve, ms));
+}
+
+async function run() {
+  console.log("  Waiting for browser to connect (up to 2 minutes)...");
+
+  for (let attempt = 1; attempt <= MAX_ATTEMPTS; attempt++) {
+    try {
+      const result = execSync(
+        `${COMPOSE} exec -T openclaw node -e "${script.replace(/"/g, '\\"')}"`,
+        { encoding: "utf8", timeout: 15000, env: EXEC_ENV }
+      ).trim();
+
+      // New device(s) approved
+      const approvedMatch = result.match(/APPROVED:(\d+)/);
+      if (approvedMatch && parseInt(approvedMatch[1]) > 0) {
+        console.log(`  Auto-approved ${approvedMatch[1]} device(s) — ready to use!`);
+        return;
+      }
+
+      // Device already paired from a previous session
+      const pairedMatch = result.match(/ALREADY_PAIRED:(\d+)/);
+      if (pairedMatch && parseInt(pairedMatch[1]) > 0) {
+        console.log(`  Device already paired (${pairedMatch[1]} device(s)) — ready to use!`);
+        return;
+      }
+
+      // Nothing yet — keep polling
+      if (attempt < MAX_ATTEMPTS) {
+        // Only log every 4th attempt to avoid spam
+        if (attempt % 4 === 0) {
+          const remaining = Math.round((MAX_ATTEMPTS - attempt) * DELAY_MS / 1000);
+          console.log(`  Still waiting for browser connection... (${remaining}s remaining)`);
+        }
+        await sleep(DELAY_MS);
+      } else {
+        console.log("  Timeout: No device connected after 2 minutes.");
+        console.log("  If you see NOT_PAIRED errors, open the browser and restart the app.");
+      }
+    } catch (e) {
+      if (attempt < MAX_ATTEMPTS) {
+        // Container might still be starting — retry silently
+        await sleep(DELAY_MS);
+      } else {
+        console.log("  Device auto-approve failed:", e.message.split("\n")[0]);
+      }
+    }
+  }
+}
+
+run();

package/cli/index.js

@@ -0,0 +1,372 @@
+#!/usr/bin/env node
+// OpenVoiceUI CLI — zero-dependency entry point
+// Usage: openvoiceui <command> [options]
+
+const { execSync, spawn } = require("child_process");
+const fs = require("fs");
+const path = require("path");
+
+const VERSION = require("../package.json").version;
+const PROJECT_DIR = path.resolve(__dirname, "..");
+
+// ---------------------------------------------------------------------------
+// Helpers
+// ---------------------------------------------------------------------------
+
+function run(cmd, opts = {}) {
+  return execSync(cmd, {
+    cwd: PROJECT_DIR,
+    stdio: opts.silent ? "pipe" : "inherit",
+    ...opts,
+  });
+}
+
+function runLive(cmd, args = []) {
+  const child = spawn(cmd, args, {
+    cwd: PROJECT_DIR,
+    stdio: "inherit",
+    shell: true,
+  });
+  child.on("error", (err) => {
+    console.error(`Failed to run: ${cmd} ${args.join(" ")}`);
+    console.error(err.message);
+    process.exit(1);
+  });
+  return child;
+}
+
+function hasDocker() {
+  try {
+    execSync("docker --version", { stdio: "pipe" });
+    execSync("docker compose version", { stdio: "pipe" });
+    return true;
+  } catch {
+    return false;
+  }
+}
+
+function isRunning() {
+  try {
+    const out = execSync("docker compose ps --format json", {
+      cwd: PROJECT_DIR,
+      stdio: "pipe",
+    }).toString();
+    // docker compose ps returns one JSON object per line
+    const lines = out.trim().split("\n").filter(Boolean);
+    return lines.length > 0;
+  } catch {
+    return false;
+  }
+}
+
+function getPort() {
+  const envPath = path.join(PROJECT_DIR, ".env");
+  if (fs.existsSync(envPath)) {
+    const content = fs.readFileSync(envPath, "utf8");
+    const match = content.match(/^PORT=(\d+)/m);
+    if (match) return match[1];
+  }
+  return "5001";
+}
+
+function printBanner() {
+  console.log(`
+  ╔══════════════════════════════════════╗
+  ║         OpenVoiceUI v${VERSION.padEnd(14)}║
+  ║   Voice-Powered AI Assistant        ║
+  ╚══════════════════════════════════════╝
+  `);
+}
+
+// ---------------------------------------------------------------------------
+// Commands
+// ---------------------------------------------------------------------------
+
+const commands = {
+  setup: {
+    desc: "Interactive setup — configure API keys and build Docker images",
+    run: cmdSetup,
+  },
+  start: {
+    desc: "Start OpenVoiceUI (Docker Compose)",
+    run: cmdStart,
+  },
+  stop: {
+    desc: "Stop OpenVoiceUI",
+    run: cmdStop,
+  },
+  restart: {
+    desc: "Restart OpenVoiceUI",
+    run: cmdRestart,
+  },
+  status: {
+    desc: "Show container status",
+    run: cmdStatus,
+  },
+  logs: {
+    desc: "Stream container logs (Ctrl+C to stop)",
+    run: cmdLogs,
+  },
+  update: {
+    desc: "Pull latest source and rebuild images",
+    run: cmdUpdate,
+  },
+  config: {
+    desc: "Open the OpenClaw control panel (localhost:18791)",
+    run: cmdConfig,
+  },
+  help: {
+    desc: "Show this help message",
+    run: cmdHelp,
+  },
+};
+
+// --- setup ---
+async function cmdSetup() {
+  printBanner();
+  console.log("  Setting up OpenVoiceUI...\n");
+
+  if (!hasDocker()) {
+    console.error("  ERROR: Docker and Docker Compose are required.");
+    console.error("  Install Docker: https://docs.docker.com/get-docker/\n");
+    process.exit(1);
+  }
+  console.log("  [OK] Docker found\n");
+
+  // Interactive key collection
+  const readline = require("readline");
+  const rl = readline.createInterface({
+    input: process.stdin,
+    output: process.stdout,
+  });
+
+  const ask = (q) =>
+    new Promise((resolve) =>
+      rl.question(q, (a) => resolve(a.trim()))
+    );
+
+  console.log("  ── Required Keys ──────────────────────────────────────");
+  console.log("  These are needed for the app to function.\n");
+
+  const groqKey = await ask(
+    "  Groq API Key (console.groq.com) [REQUIRED]: "
+  );
+  if (!groqKey) {
+    console.error("\n  ERROR: Groq API key is required for TTS.\n");
+    rl.close();
+    process.exit(1);
+  }
+
+  const deepgramKey = await ask(
+    "  Deepgram API Key (console.deepgram.com) [REQUIRED]: "
+  );
+  if (!deepgramKey) {
+    console.error("\n  ERROR: Deepgram API key is required for STT.\n");
+    rl.close();
+    process.exit(1);
+  }
+
+  console.log("\n  ── AI Provider Keys (pick at least one) ──────────────");
+  console.log("  Press Enter to skip any provider.\n");
+
+  const zaiKey = await ask("  Z.AI API Key (z.ai): ");
+  const anthropicKey = await ask("  Anthropic API Key (console.anthropic.com): ");
+  const openaiKey = await ask("  OpenAI API Key (platform.openai.com): ");
+
+  if (!zaiKey && !anthropicKey && !openaiKey) {
+    console.log(
+      "\n  WARNING: No AI provider key set. You can add one later in the .env file.\n"
+    );
+  }
+
+  console.log("\n  ── Optional Keys ─────────────────────────────────────");
+  console.log("  Press Enter to skip.\n");
+
+  const geminiKey = await ask("  Gemini API Key (aistudio.google.com): ");
+  const openrouterKey = await ask("  OpenRouter API Key (openrouter.ai): ");
+  const sunoKey = await ask("  Suno API Key — AI Music (sunoapi.org): ");
+
+  const portAnswer = await ask("  Port (default 5001): ");
+  const port = portAnswer || "5001";
+
+  rl.close();
+
+  // Set env vars for setup-config.js
+  const setupEnv = {
+    PINOKIO_PORT: port,
+    PINOKIO_GROQ_API_KEY: groqKey,
+    PINOKIO_DEEPGRAM_API_KEY: deepgramKey,
+    PINOKIO_ZAI_API_KEY: zaiKey,
+    PINOKIO_ANTHROPIC_API_KEY: anthropicKey,
+    PINOKIO_OPENAI_API_KEY: openaiKey,
+    PINOKIO_GEMINI_API_KEY: geminiKey,
+    PINOKIO_OPENROUTER_API_KEY: openrouterKey,
+    PINOKIO_SUNO_API_KEY: sunoKey,
+  };
+
+  console.log("\n  Generating configuration files...\n");
+  run("node setup-config.js", {
+    env: { ...process.env, ...setupEnv },
+  });
+
+  console.log("\n  Building Docker images (this may take a few minutes)...\n");
+  run("docker compose build");
+
+  console.log(`
+  ════════════════════════════════════════════════════════
+  Setup complete!
+
+  Start OpenVoiceUI:   openvoiceui start
+  Open in browser:     http://localhost:${port}
+  OpenClaw control:    http://localhost:18791
+  View logs:           openvoiceui logs
+  ════════════════════════════════════════════════════════
+  `);
+}
+
+// --- start ---
+function cmdStart() {
+  printBanner();
+  if (!hasDocker()) {
+    console.error("  ERROR: Docker not found. Run: openvoiceui setup\n");
+    process.exit(1);
+  }
+
+  const envPath = path.join(PROJECT_DIR, ".env");
+  if (!fs.existsSync(envPath)) {
+    console.error("  ERROR: No .env file found. Run: openvoiceui setup\n");
+    process.exit(1);
+  }
+
+  console.log("  Starting OpenVoiceUI...\n");
+  run("docker compose up -d");
+
+  // Inject pre-paired device identity if available
+  const prePairedPath = path.join(
+    PROJECT_DIR,
+    "openclaw-data",
+    "pre-paired-device.json"
+  );
+  if (fs.existsSync(prePairedPath)) {
+    try {
+      run("node inject-device-identity.js", { silent: true });
+    } catch {
+      // Non-fatal — device pairing may already be done
+    }
+  }
+
+  const port = getPort();
+  console.log(`
+  OpenVoiceUI is running!
+
+  App:          http://localhost:${port}
+  OpenClaw:     http://localhost:18791
+  Stop:         openvoiceui stop
+  Logs:         openvoiceui logs
+  `);
+}
+
+// --- stop ---
+function cmdStop() {
+  console.log("  Stopping OpenVoiceUI...\n");
+  run("docker compose down");
+  console.log("  Stopped.\n");
+}
+
+// --- restart ---
+function cmdRestart() {
+  console.log("  Restarting OpenVoiceUI...\n");
+  run("docker compose down");
+  run("docker compose up -d");
+
+  const port = getPort();
+  console.log(`\n  Restarted. App: http://localhost:${port}\n`);
+}
+
+// --- status ---
+function cmdStatus() {
+  printBanner();
+  if (!hasDocker()) {
+    console.error("  Docker not found.\n");
+    process.exit(1);
+  }
+  run("docker compose ps");
+}
+
+// --- logs ---
+function cmdLogs() {
+  const child = runLive("docker", ["compose", "logs", "-f", "--tail", "100"]);
+  process.on("SIGINT", () => {
+    child.kill("SIGINT");
+    process.exit(0);
+  });
+}
+
+// --- update ---
+function cmdUpdate() {
+  printBanner();
+  console.log("  Updating OpenVoiceUI...\n");
+
+  // Check if this is a git repo (cloned install) or npm install
+  const gitDir = path.join(PROJECT_DIR, ".git");
+  if (fs.existsSync(gitDir)) {
+    console.log("  Pulling latest source...\n");
+    run("git pull --ff-only");
+  } else {
+    console.log("  Installed via npm — update with: npm update -g openvoiceui\n");
+  }
+
+  console.log("  Rebuilding Docker images...\n");
+  run("docker compose build --pull");
+
+  if (isRunning()) {
+    console.log("  Restarting containers...\n");
+    run("docker compose up -d");
+  }
+
+  console.log("  Update complete!\n");
+}
+
+// --- config ---
+function cmdConfig() {
+  console.log("  OpenClaw control panel: http://localhost:18791\n");
+  console.log("  Open this URL in your browser to change AI models,");
+  console.log("  add provider keys, and configure agent behavior.\n");
+}
+
+// --- help ---
+function cmdHelp() {
+  printBanner();
+  console.log("  Usage: openvoiceui <command>\n");
+  console.log("  Commands:\n");
+  for (const [name, cmd] of Object.entries(commands)) {
+    console.log(`    ${name.padEnd(12)} ${cmd.desc}`);
+  }
+  console.log(`
+  Quick start:
+    openvoiceui setup     # configure keys + build images
+    openvoiceui start     # launch the app
+    open http://localhost:5001
+  `);
+}
+
+// ---------------------------------------------------------------------------
+// Main
+// ---------------------------------------------------------------------------
+
+const command = process.argv[2] || "help";
+
+if (commands[command]) {
+  const result = commands[command].run();
+  // Handle async commands (setup uses readline)
+  if (result && typeof result.catch === "function") {
+    result.catch((err) => {
+      console.error(`\n  ERROR: ${err.message}\n`);
+      process.exit(1);
+    });
+  }
+} else {
+  console.error(`  Unknown command: ${command}\n`);
+  cmdHelp();
+  process.exit(1);
+}

package/config/__init__.py

@@ -0,0 +1,4 @@
+# config package
+from config.loader import config
+
+__all__ = ["config"]

package/config/default.yaml

@@ -0,0 +1,43 @@
+# OpenVoiceUI Default Configuration
+# All values here can be overridden by environment variables.
+# Env var naming: SECTION__KEY (double underscore), e.g. SERVER__PORT=5002
+# Or use the direct env var names listed in comments.
+
+# ── Server ────────────────────────────────────────────────────────────────────
+server:
+  host: "0.0.0.0"
+  port: 5001                # env: PORT
+  debug: false
+  threaded: true
+
+# ── Gateway (OpenClaw) ────────────────────────────────────────────────────────
+gateway:
+  url: "ws://127.0.0.1:18791"         # env: CLAWDBOT_GATEWAY_URL
+  auth_token: ""                       # env: CLAWDBOT_AUTH_TOKEN (required)
+  session_key: "voice-main-1"          # env: GATEWAY_SESSION_KEY
+
+# ── LLM Models ────────────────────────────────────────────────────────────────
+models:
+  whisper: "tiny"                      # Faster-Whisper model size
+  whisper_device: "cpu"                # cpu or cuda
+  whisper_compute: "float32"           # float32, float16, int8
+
+# ── TTS ───────────────────────────────────────────────────────────────────────
+tts:
+  provider: "groq"                     # env: TTS_PROVIDER (groq or supertonic)
+
+# ── Conversation ──────────────────────────────────────────────────────────────
+conversation:
+  max_history_messages: 20             # env: MAX_HISTORY_MESSAGES
+  active_profile: "default"            # must match a filename in profiles/
+
+# ── Logging ───────────────────────────────────────────────────────────────────
+logging:
+  level: "INFO"                        # env: LOG_LEVEL
+
+# ── Features (runtime toggles) ────────────────────────────────────────────────
+features:
+  enable_fts: true                     # env: ENABLE_FTS
+  enable_briefing: true                # env: ENABLE_BRIEFING
+  enable_history_reload: true          # env: ENABLE_HISTORY_RELOAD
+  memory_client: true                  # whether to use memory_client if available

package/config/flags.yaml

@@ -0,0 +1,67 @@
+# Feature Flags — ai-eyes2 refactor
+#
+# These flags gate each refactor phase so changes can be rolled out gradually.
+# All flags default to false (old behaviour preserved until explicitly enabled).
+#
+# Override any flag via env var:
+#   FEATURE_USE_BLUEPRINTS=true
+#   FEATURE_PERSISTENT_WEBSOCKET=true
+#   etc.
+#
+# Usage in code:
+#   from config.loader import config
+#   if config.flag('use_blueprints'):
+#       ...
+
+flags:
+  # ── Phase 1: Foundation ───────────────────────────────────────────────────
+  use_yaml_config: true          # P1-T2 — YAML config system active
+  use_feature_flags: true        # P1-T3 — this flag system itself (meta)
+  use_health_probes: true        # P1-T4 — /health/live and /health/ready endpoints
+  use_sqlite_wal: true           # P1-T5 — WAL mode + connection pooling
+  use_session_reset: false       # P1-T7 — POST /api/session/reset endpoint
+
+  # ── Phase 2: Backend Blueprint Split ─────────────────────────────────────
+  use_blueprints: false          # P2-T1 — Flask app factory + blueprints
+  use_gateway_service: false     # P2-T2 — GatewayConnection as services/gateway.py
+  use_conversation_blueprint: false  # P2-T3 — conversation routes in blueprint
+  use_music_blueprint: false     # P2-T4 — music routes in blueprint
+  use_canvas_blueprint: false    # P2-T5 — canvas routes in blueprint
+  use_admin_blueprint: false     # P2-T6 — admin routes in blueprint
+  use_tts_service: false         # P2-T7 — TTS providers as service module
+
+  # ── Phase G: Gateway & Transport ─────────────────────────────────────────
+  use_persistent_websocket: false  # PG-T2 — persistent WS to Gateway
+  use_streaming_events: false      # PG-T3 — stream delta events to frontend
+  use_webchat_mode: true           # PG-T5 — Gateway client mode WEBCHAT
+
+  # ── Phase 3: Frontend Module Extraction ──────────────────────────────────
+  use_eventbus: false            # P3-T1 — frontend EventBus pub/sub
+  use_css_modules: false         # P3-T2 — extracted CSS files
+  use_stt_module: false          # P3-T3 — WebSpeechSTT as module
+  use_tts_player_module: false   # P3-T4 — TTSPlayer as module
+  use_face_module: false         # P3-T5 — face system as module
+  use_music_module: false        # P3-T6 — MusicPlayer as module
+  use_voice_session: false       # P3-T7 — VoiceSession orchestrator
+  use_emotion_engine: false      # P3-T8 — EmotionEngine for mood → face
+  use_thin_shell: false          # P3-T9 — slim index.html shell
+
+  # ── Phase 4: Admin UI ────────────────────────────────────────────────────
+  use_session_control_ui: false  # P4-T1 — Reset Conversation button
+  use_playlist_editor: false     # P4-T2 — PlaylistEditor UI
+  use_face_picker: false         # P4-T3 — face picker gallery
+  use_theme_editor: false        # P4-T4 — color picker / theme editor
+  use_tts_preview: false         # P4-T5 — TTS voice preview
+  use_profile_switcher_ui: false # P4-T6 — profile switcher UI
+
+  # ── Phase 5: Agent Profiles & Provider Registry ───────────────────────────
+  use_provider_registry: false   # P5-T2 — provider auto-discovery
+  use_profile_system: false      # P5-T3 — JSON profile schema
+  use_profile_manager: false     # P5-T4 — ProfileManager + API
+  use_speech_normalization: false # P5-T5 — config-driven speech normalisation
+
+  # ── Phase 6: Multi-Agent Framework ───────────────────────────────────────
+  use_eventbridge: false         # P6-T1 — EventBridge implementation
+  use_clawdbot_adapter: false    # P6-T2 — ClawdBotAdapter module
+  use_hume_adapter: false        # P6-T3 — Hume EVI adapter
+  use_capability_ui: false       # P6-T4 — capability-driven UI shell