Initial commit from agent

utils/audio_utils.py

@@ -0,0 +1,49 @@
+"""
+Audio utilities — microphone capture helpers using PyAudio.
+Captures audio from the default input device and returns WAV bytes.
+"""
+import io
+import wave
+import pyaudio
+
+CHUNK = 1024
+FORMAT = pyaudio.paInt16
+CHANNELS = 1
+RATE = 16000
+
+def record_audio(duration: float = 5.0) -> bytes:
+    """
+    Record audio from default microphone for the given duration.
+    Returns raw WAV bytes compatible with Whisper.
+    """
+    p = pyaudio.PyAudio()
+    stream = p.open(
+        format=FORMAT, channels=CHANNELS,
+        rate=RATE, input=True, frames_per_buffer=CHUNK
+    )
+    frames = []
+    for _ in range(int(RATE / CHUNK * duration)):
+        data = stream.read(CHUNK, exception_on_overflow=False)
+        frames.append(data)
+    stream.stop_stream()
+    stream.close()
+    p.terminate()
+
+    buf = io.BytesIO()
+    with wave.open(buf, 'wb') as wf:
+        wf.setnchannels(CHANNELS)
+        wf.setsampwidth(p.get_sample_size(FORMAT))
+        wf.setframerate(RATE)
+        wf.writeframes(b''.join(frames))
+    return buf.getvalue()
+
+def list_audio_devices() -> list[dict]:
+    """List all available audio input devices."""
+    p = pyaudio.PyAudio()
+    devices = []
+    for i in range(p.get_device_count()):
+        info = p.get_device_info_by_index(i)
+        if info['maxInputChannels'] > 0:
+            devices.append({'index': i, 'name': info['name']})
+    p.terminate()
+    return devices

utils/window_utils.py

@@ -0,0 +1,43 @@
+"""
+Windows window utilities using win32gui.
+Enumerate and manage application windows.
+"""
+try:
+    import win32gui
+    import win32con
+
+    def get_active_window() -> dict:
+        hwnd = win32gui.GetForegroundWindow()
+        return {
+            'hwnd': hwnd,
+            'title': win32gui.GetWindowText(hwnd)
+        }
+
+    def list_windows() -> list[dict]:
+        windows = []
+        def _enum(hwnd, _):
+            if win32gui.IsWindowVisible(hwnd):
+                title = win32gui.GetWindowText(hwnd)
+                if title:
+                    windows.append({'hwnd': hwnd, 'title': title})
+        win32gui.EnumWindows(_enum, None)
+        return windows
+
+    def focus_window(title_substring: str) -> bool:
+        """Bring a window to foreground by partial title match."""
+        def _enum(hwnd, _):
+            if title_substring.lower() in win32gui.GetWindowText(hwnd).lower():
+                win32gui.SetForegroundWindow(hwnd)
+                return False
+        win32gui.EnumWindows(_enum, None)
+        return True
+
+except ImportError:
+    def get_active_window():
+        return {'hwnd': None, 'title': 'Unknown (pywin32 not installed)'}
+
+    def list_windows():
+        return []
+
+    def focus_window(title_substring: str):
+        return False