agent-cli 0.2.0__tar.gz → 0.3.1__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (29) hide show
  1. {agent_cli-0.2.0 → agent_cli-0.3.1}/PKG-INFO +56 -33
  2. {agent_cli-0.2.0 → agent_cli-0.3.1}/README.md +55 -32
  3. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/agents/_cli_options.py +10 -5
  4. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/agents/autocorrect.py +22 -26
  5. agent_cli-0.3.1/agent_cli/agents/transcribe.py +220 -0
  6. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/agents/voice_assistant.py +31 -113
  7. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/asr.py +8 -7
  8. agent_cli-0.3.1/agent_cli/llm.py +135 -0
  9. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/utils.py +62 -0
  10. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli.egg-info/PKG-INFO +56 -33
  11. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli.egg-info/SOURCES.txt +2 -2
  12. agent_cli-0.2.0/tests/test_ollama_client.py → agent_cli-0.3.1/tests/test_llm.py +1 -1
  13. agent_cli-0.2.0/agent_cli/agents/transcribe.py +0 -131
  14. agent_cli-0.2.0/agent_cli/ollama_client.py +0 -24
  15. {agent_cli-0.2.0 → agent_cli-0.3.1}/LICENSE +0 -0
  16. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/__init__.py +0 -0
  17. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/agents/__init__.py +0 -0
  18. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/cli.py +0 -0
  19. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/config.py +0 -0
  20. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/process_manager.py +0 -0
  21. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/py.typed +0 -0
  22. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli.egg-info/dependency_links.txt +0 -0
  23. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli.egg-info/entry_points.txt +0 -0
  24. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli.egg-info/requires.txt +0 -0
  25. {agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli.egg-info/top_level.txt +0 -0
  26. {agent_cli-0.2.0 → agent_cli-0.3.1}/pyproject.toml +0 -0
  27. {agent_cli-0.2.0 → agent_cli-0.3.1}/setup.cfg +0 -0
  28. {agent_cli-0.2.0 → agent_cli-0.3.1}/tests/test_process_manager.py +0 -0
  29. {agent_cli-0.2.0 → agent_cli-0.3.1}/tests/test_utils.py +0 -0
{agent_cli-0.2.0 → agent_cli-0.3.1}/PKG-INFO RENAMED
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: agent-cli
3
- Version: 0.2.0
3
+ Version: 0.3.1
4
4
  Summary: A suite of AI-powered command-line tools for text correction, audio transcription, and voice assistance.
5
5
  Author-email: Bas Nijholt <bas@nijho.lt>
6
6
  Project-URL: Homepage, https://github.com/basnijholt/agent-cli
@@ -196,13 +196,23 @@ Transcribes whatever you say into text using Wyoming ASR (Automatic Speech Recog
196
196
  │ [default: 192.168.1.143] │
197
197
  │ --asr-server-port INTEGER Wyoming ASR server port. │
198
198
  │ [default: 10300] │
199
- │ --clipboard --no-clipboard Copy transcript to │
200
- │ clipboard. │
201
- │ [default: clipboard] │
199
+ │ --model -m TEXT The Ollama model to use. │
200
+ │ Default is devstral:24b. │
201
+ │ [default: devstral:24b] │
202
+ │ --ollama-host TEXT The Ollama server host. │
203
+ │ Default is │
204
+ │ http://localhost:11434. │
205
+ │ [default: │
206
+ │ http://localhost:11434] │
207
+ │ --llm --no-llm Use an LLM to process the │
208
+ │ transcript. │
209
+ │ [default: no-llm] │
202
210
  │ --stop Stop any running │
203
211
  │ background process. │
204
212
  │ --status Check if a background │
205
213
  │ process is running. │
214
+ │ --clipboard --no-clipboard Copy result to clipboard. │
215
+ │ [default: clipboard] │
206
216
  │ --log-level TEXT Set logging level. │
207
217
  │ [default: WARNING] │
208
218
  │ --log-file TEXT Path to a file to write │
@@ -268,35 +278,48 @@ The process management features make it perfect for hotkey toggles. Use `--statu
268
278
  voice-assistant --stop
269
279
 
270
280
  ╭─ Options ────────────────────────────────────────────────────────────────────╮
271
- │ --device-index INTEGER Index of the PyAudio input device to │
272
- │ use. │
273
- │ [default: None] │
274
- │ --device-name TEXT Device name keywords for partial │
275
- │ matching. Supports comma-separated list │
276
- │ where each term can partially match │
277
- │ device names (case-insensitive). First │
278
- │ matching device is selected. │
279
- │ [default: None] │
280
- │ --list-devices List available audio input devices and │
281
- │ exit. │
282
- │ --asr-server-ip TEXT Wyoming ASR server IP address. │
283
- │ [default: 192.168.1.143] │
284
- │ --asr-server-port INTEGER Wyoming ASR server port. │
285
- │ [default: 10300] │
286
- │ --model -m TEXT The Ollama model to use. Default is │
287
- │ devstral:24b. │
288
- │ [default: devstral:24b] │
289
- │ --ollama-host TEXT The Ollama server host. Default is │
290
- │ http://localhost:11434. │
291
- │ [default: http://localhost:11434] │
292
- │ --stop Stop any running background process. │
293
- │ --status Check if a background process is │
294
- │ running. │
295
- │ --log-level TEXT Set logging level. [default: WARNING] │
296
- │ --log-file TEXT Path to a file to write logs to. │
297
- │ [default: None] │
298
- │ --quiet -q Suppress console output from rich. │
299
- │ --help Show this message and exit. │
281
+ │ --device-index INTEGER Index of the PyAudio input │
282
+ │ device to use. │
283
+ │ [default: None] │
284
+ │ --device-name TEXT Device name keywords for │
285
+ │ partial matching. Supports │
286
+ │ comma-separated list where │
287
+ │ each term can partially │
288
+ │ match device names │
289
+ │ (case-insensitive). First │
290
+ │ matching device is │
291
+ │ selected. │
292
+ │ [default: None] │
293
+ │ --list-devices List available audio input │
294
+ │ devices and exit. │
295
+ │ --asr-server-ip TEXT Wyoming ASR server IP │
296
+ │ address. │
297
+ │ [default: 192.168.1.143] │
298
+ │ --asr-server-port INTEGER Wyoming ASR server port. │
299
+ │ [default: 10300] │
300
+ │ --model -m TEXT The Ollama model to use. │
301
+ │ Default is devstral:24b. │
302
+ │ [default: devstral:24b] │
303
+ │ --ollama-host TEXT The Ollama server host. │
304
+ │ Default is │
305
+ │ http://localhost:11434. │
306
+ │ [default: │
307
+ │ http://localhost:11434] │
308
+ │ --stop Stop any running │
309
+ │ background process. │
310
+ │ --status Check if a background │
311
+ │ process is running. │
312
+ │ --clipboard --no-clipboard Copy result to clipboard. │
313
+ │ [default: clipboard] │
314
+ │ --log-level TEXT Set logging level. │
315
+ │ [default: WARNING] │
316
+ │ --log-file TEXT Path to a file to write │
317
+ │ logs to. │
318
+ │ [default: None] │
319
+ │ --quiet -q Suppress console output │
320
+ │ from rich. │
321
+ │ --help Show this message and │
322
+ │ exit. │
300
323
  ╰──────────────────────────────────────────────────────────────────────────────╯
301
324
 
302
325
  ```
{agent_cli-0.2.0 → agent_cli-0.3.1}/README.md RENAMED
@@ -166,13 +166,23 @@ Transcribes whatever you say into text using Wyoming ASR (Automatic Speech Recog
166
166
  │ [default: 192.168.1.143] │
167
167
  │ --asr-server-port INTEGER Wyoming ASR server port. │
168
168
  │ [default: 10300] │
169
- │ --clipboard --no-clipboard Copy transcript to │
170
- │ clipboard. │
171
- │ [default: clipboard] │
169
+ │ --model -m TEXT The Ollama model to use. │
170
+ │ Default is devstral:24b. │
171
+ │ [default: devstral:24b] │
172
+ │ --ollama-host TEXT The Ollama server host. │
173
+ │ Default is │
174
+ │ http://localhost:11434. │
175
+ │ [default: │
176
+ │ http://localhost:11434] │
177
+ │ --llm --no-llm Use an LLM to process the │
178
+ │ transcript. │
179
+ │ [default: no-llm] │
172
180
  │ --stop Stop any running │
173
181
  │ background process. │
174
182
  │ --status Check if a background │
175
183
  │ process is running. │
184
+ │ --clipboard --no-clipboard Copy result to clipboard. │
185
+ │ [default: clipboard] │
176
186
  │ --log-level TEXT Set logging level. │
177
187
  │ [default: WARNING] │
178
188
  │ --log-file TEXT Path to a file to write │
@@ -238,35 +248,48 @@ The process management features make it perfect for hotkey toggles. Use `--statu
238
248
  voice-assistant --stop
239
249
 
240
250
  ╭─ Options ────────────────────────────────────────────────────────────────────╮
241
- │ --device-index INTEGER Index of the PyAudio input device to │
242
- │ use. │
243
- │ [default: None] │
244
- │ --device-name TEXT Device name keywords for partial │
245
- │ matching. Supports comma-separated list │
246
- │ where each term can partially match │
247
- │ device names (case-insensitive). First │
248
- │ matching device is selected. │
249
- │ [default: None] │
250
- │ --list-devices List available audio input devices and │
251
- │ exit. │
252
- │ --asr-server-ip TEXT Wyoming ASR server IP address. │
253
- │ [default: 192.168.1.143] │
254
- │ --asr-server-port INTEGER Wyoming ASR server port. │
255
- │ [default: 10300] │
256
- │ --model -m TEXT The Ollama model to use. Default is │
257
- │ devstral:24b. │
258
- │ [default: devstral:24b] │
259
- │ --ollama-host TEXT The Ollama server host. Default is │
260
- │ http://localhost:11434. │
261
- │ [default: http://localhost:11434] │
262
- │ --stop Stop any running background process. │
263
- │ --status Check if a background process is │
264
- │ running. │
265
- │ --log-level TEXT Set logging level. [default: WARNING] │
266
- │ --log-file TEXT Path to a file to write logs to. │
267
- │ [default: None] │
268
- │ --quiet -q Suppress console output from rich. │
269
- │ --help Show this message and exit. │
251
+ │ --device-index INTEGER Index of the PyAudio input │
252
+ │ device to use. │
253
+ │ [default: None] │
254
+ │ --device-name TEXT Device name keywords for │
255
+ │ partial matching. Supports │
256
+ │ comma-separated list where │
257
+ │ each term can partially │
258
+ │ match device names │
259
+ │ (case-insensitive). First │
260
+ │ matching device is │
261
+ │ selected. │
262
+ │ [default: None] │
263
+ │ --list-devices List available audio input │
264
+ │ devices and exit. │
265
+ │ --asr-server-ip TEXT Wyoming ASR server IP │
266
+ │ address. │
267
+ │ [default: 192.168.1.143] │
268
+ │ --asr-server-port INTEGER Wyoming ASR server port. │
269
+ │ [default: 10300] │
270
+ │ --model -m TEXT The Ollama model to use. │
271
+ │ Default is devstral:24b. │
272
+ │ [default: devstral:24b] │
273
+ │ --ollama-host TEXT The Ollama server host. │
274
+ │ Default is │
275
+ │ http://localhost:11434. │
276
+ │ [default: │
277
+ │ http://localhost:11434] │
278
+ │ --stop Stop any running │
279
+ │ background process. │
280
+ │ --status Check if a background │
281
+ │ process is running. │
282
+ │ --clipboard --no-clipboard Copy result to clipboard. │
283
+ │ [default: clipboard] │
284
+ │ --log-level TEXT Set logging level. │
285
+ │ [default: WARNING] │
286
+ │ --log-file TEXT Path to a file to write │
287
+ │ logs to. │
288
+ │ [default: None] │
289
+ │ --quiet -q Suppress console output │
290
+ │ from rich. │
291
+ │ --help Show this message and │
292
+ │ exit. │
270
293
  ╰──────────────────────────────────────────────────────────────────────────────╯
271
294
 
272
295
  ```
{agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/agents/_cli_options.py RENAMED
@@ -18,6 +18,11 @@ OLLAMA_HOST: str = typer.Option(
18
18
  "--ollama-host",
19
19
  help=f"The Ollama server host. Default is {config.OLLAMA_HOST}.",
20
20
  )
21
+ LLM: bool = typer.Option(
22
+ False, # noqa: FBT003
23
+ "--llm/--no-llm",
24
+ help="Use an LLM to process the transcript.",
25
+ )
21
26
 
22
27
 
23
28
  # --- ASR (Audio) Options ---
@@ -47,11 +52,6 @@ ASR_SERVER_PORT: int = typer.Option(
47
52
  "--asr-server-port",
48
53
  help="Wyoming ASR server port.",
49
54
  )
50
- CLIPBOARD: bool = typer.Option(
51
- True, # noqa: FBT003
52
- "--clipboard/--no-clipboard",
53
- help="Copy transcript to clipboard.",
54
- )
55
55
 
56
56
 
57
57
  # --- Process Management Options ---
@@ -68,6 +68,11 @@ STATUS: bool = typer.Option(
68
68
 
69
69
 
70
70
  # --- General Options ---
71
+ CLIPBOARD: bool = typer.Option(
72
+ True, # noqa: FBT003
73
+ "--clipboard/--no-clipboard",
74
+ help="Copy result to clipboard.",
75
+ )
71
76
  LOG_LEVEL: str = typer.Option(
72
77
  "WARNING",
73
78
  "--log-level",
{agent_cli-0.2.0 → agent_cli-0.3.1}/agent_cli/agents/autocorrect.py RENAMED
@@ -27,13 +27,18 @@ import typer
27
27
  from openai import APIConnectionError
28
28
  from pydantic_ai.exceptions import ModelHTTPError
29
29
  from rich.console import Console
30
- from rich.panel import Panel
31
30
  from rich.status import Status
32
31
 
33
32
  import agent_cli.agents._cli_options as opts
34
33
  from agent_cli.cli import app, setup_logging
35
- from agent_cli.ollama_client import build_agent
36
- from agent_cli.utils import get_clipboard_text
34
+ from agent_cli.llm import build_agent
35
+ from agent_cli.utils import (
36
+ get_clipboard_text,
37
+ print_error_message,
38
+ print_input_panel,
39
+ print_output_panel,
40
+ print_status_message,
41
+ )
37
42
 
38
43
  # --- Configuration ---
39
44
 
@@ -71,16 +76,7 @@ async def process_text(text: str, model: str, ollama_host: str) -> tuple[str, fl
71
76
 
72
77
  def display_original_text(original_text: str, console: Console | None) -> None:
73
78
  """Render the original text panel in verbose mode."""
74
- if console is None:
75
- return
76
- console.print(
77
- Panel(
78
- original_text,
79
- title="[bold cyan]📋 Original Text[/bold cyan]",
80
- border_style="cyan",
81
- padding=(1, 2),
82
- ),
83
- )
79
+ print_input_panel(console, original_text, title="📋 Original Text")
84
80
 
85
81
 
86
82
  def _display_result(
@@ -101,16 +97,15 @@ def _display_result(
101
97
  print(corrected_text)
102
98
  else:
103
99
  assert console is not None
104
- console.print(
105
- Panel(
106
- corrected_text,
107
- title="[bold green]✨ Corrected Text[/bold green]",
108
- border_style="green",
109
- padding=(1, 2),
110
- ),
100
+ print_output_panel(
101
+ console,
102
+ corrected_text,
103
+ title="✨ Corrected Text",
104
+ subtitle=f"[dim]took {elapsed:.2f}s[/dim]",
111
105
  )
112
- console.print(
113
- f"✅ [bold green]Success! Corrected text has been copied to your clipboard. [bold yellow](took {elapsed:.2f} seconds)[/bold yellow][/bold green]",
106
+ print_status_message(
107
+ console,
108
+ "✅ Success! Corrected text has been copied to your clipboard.",
114
109
  )
115
110
 
116
111
 
@@ -166,9 +161,10 @@ def autocorrect(
166
161
  except (httpx.ConnectError, ModelHTTPError, APIConnectionError) as e:
167
162
  if quiet:
168
163
  print(f"❌ {e}")
169
- elif console:
170
- console.print(f"❌ {e}", style="bold red")
171
- console.print(
172
- f" Please check that your Ollama server is running at [bold cyan]{ollama_host}[/bold cyan]",
164
+ else:
165
+ print_error_message(
166
+ console,
167
+ str(e),
168
+ f"Please check that your Ollama server is running at [bold cyan]{ollama_host}[/bold cyan]",
173
169
  )
174
170
  sys.exit(1)
agent_cli-0.3.1/agent_cli/agents/transcribe.py ADDED
@@ -0,0 +1,220 @@
1
+ """Wyoming ASR Client for streaming microphone audio to a transcription server."""
2
+
3
+ from __future__ import annotations
4
+
5
+ import asyncio
6
+ import logging
7
+ from contextlib import AbstractContextManager, nullcontext, suppress
8
+ from typing import TYPE_CHECKING
9
+
10
+ import pyperclip
11
+ from rich.console import Console
12
+ from rich.live import Live
13
+ from rich.text import Text
14
+
15
+ import agent_cli.agents._cli_options as opts
16
+ from agent_cli import asr, process_manager
17
+ from agent_cli.cli import app, setup_logging
18
+ from agent_cli.llm import process_and_update_clipboard
19
+ from agent_cli.utils import (
20
+ print_device_index,
21
+ print_input_panel,
22
+ print_output_panel,
23
+ print_status_message,
24
+ signal_handling_context,
25
+ )
26
+
27
+ if TYPE_CHECKING:
28
+ import pyaudio
29
+
30
+ LOGGER = logging.getLogger()
31
+
32
+ SYSTEM_PROMPT = """
33
+ You are an AI transcription cleanup assistant. Your purpose is to improve and refine raw speech-to-text transcriptions by correcting errors, adding proper punctuation, and enhancing readability while preserving the original meaning and intent.
34
+
35
+ Your tasks include:
36
+ - Correcting obvious speech recognition errors and mishearing
37
+ - Adding appropriate punctuation (periods, commas, question marks, etc.)
38
+ - Fixing capitalization where needed
39
+ - Removing filler words, false starts, and repeated words when they clearly weren't intentional
40
+ - Improving sentence structure and flow while maintaining the speaker's voice and meaning
41
+ - Formatting the text for better readability
42
+
43
+ Important rules:
44
+ - Do not change the core meaning or content of the transcription
45
+ - Do not add information that wasn't spoken
46
+ - Do not remove content unless it's clearly an error or filler
47
+ - Return ONLY the cleaned-up text without any explanations or commentary
48
+ - Do not wrap your output in markdown or code blocks
49
+ """
50
+
51
+ AGENT_INSTRUCTIONS = """
52
+ You will be given a block of raw transcribed text enclosed in <original-text> tags, and a cleanup instruction enclosed in <instruction> tags.
53
+
54
+ Your job is to process the transcribed text according to the instruction, which will typically involve:
55
+ - Correcting speech recognition errors
56
+ - Adding proper punctuation and capitalization
57
+ - Removing obvious filler words and false starts
58
+ - Improving readability while preserving meaning
59
+
60
+ Return ONLY the cleaned-up text with no additional formatting or commentary.
61
+ """
62
+
63
+ INSTRUCTION = """
64
+ Please clean up this transcribed text by correcting any speech recognition errors, adding appropriate punctuation and capitalization, removing obvious filler words or false starts, and improving overall readability while preserving the original meaning and intent of the speaker.
65
+ """
66
+
67
+
68
+ async def async_main(
69
+ *,
70
+ device_index: int | None,
71
+ asr_server_ip: str,
72
+ asr_server_port: int,
73
+ clipboard: bool,
74
+ quiet: bool,
75
+ model: str,
76
+ ollama_host: str,
77
+ llm: bool,
78
+ console: Console | None,
79
+ p: pyaudio.PyAudio,
80
+ ) -> None:
81
+ """Async entry point, consuming parsed args."""
82
+ with (
83
+ signal_handling_context(console, LOGGER) as stop_event,
84
+ _maybe_live(console) as live,
85
+ ):
86
+ transcript = await asr.transcribe_audio(
87
+ asr_server_ip=asr_server_ip,
88
+ asr_server_port=asr_server_port,
89
+ device_index=device_index,
90
+ logger=LOGGER,
91
+ p=p,
92
+ stop_event=stop_event,
93
+ console=console,
94
+ live=live,
95
+ listening_message="Listening...",
96
+ )
97
+
98
+ if llm and model and ollama_host and transcript:
99
+ print_input_panel(console, transcript, title="📝 Raw Transcript")
100
+ await process_and_update_clipboard(
101
+ system_prompt=SYSTEM_PROMPT,
102
+ agent_instructions=AGENT_INSTRUCTIONS,
103
+ model=model,
104
+ ollama_host=ollama_host,
105
+ logger=LOGGER,
106
+ console=console,
107
+ original_text=transcript,
108
+ instruction=INSTRUCTION,
109
+ clipboard=clipboard,
110
+ )
111
+ return
112
+
113
+ # When not using LLM, show transcript in output panel for consistency
114
+ if transcript:
115
+ if quiet:
116
+ # Quiet mode: print result to stdout for Keyboard Maestro to capture
117
+ print(transcript)
118
+ else:
119
+ print_output_panel(
120
+ console,
121
+ transcript,
122
+ title="📝 Transcript",
123
+ subtitle="[dim]Copied to clipboard[/dim]" if clipboard else None,
124
+ )
125
+
126
+ if clipboard:
127
+ pyperclip.copy(transcript)
128
+ LOGGER.info("Copied transcript to clipboard.")
129
+ else:
130
+ LOGGER.info("Clipboard copy disabled.")
131
+ else:
132
+ LOGGER.info("Transcript empty.")
133
+ if not quiet:
134
+ print_status_message(console, "⚠️ No transcript captured.", style="yellow")
135
+
136
+
137
+ def _maybe_live(console: Console | None) -> AbstractContextManager[Live | None]:
138
+ if console:
139
+ return Live(
140
+ Text("Transcribing...", style="blue"),
141
+ console=console,
142
+ transient=True,
143
+ )
144
+ return nullcontext()
145
+
146
+
147
+ @app.command("transcribe")
148
+ def transcribe(
149
+ *,
150
+ device_index: int | None = opts.DEVICE_INDEX,
151
+ device_name: str | None = opts.DEVICE_NAME,
152
+ # ASR
153
+ list_devices: bool = opts.LIST_DEVICES,
154
+ asr_server_ip: str = opts.ASR_SERVER_IP,
155
+ asr_server_port: int = opts.ASR_SERVER_PORT,
156
+ # LLM
157
+ model: str = opts.MODEL,
158
+ ollama_host: str = opts.OLLAMA_HOST,
159
+ llm: bool = opts.LLM,
160
+ # Process control
161
+ stop: bool = opts.STOP,
162
+ status: bool = opts.STATUS,
163
+ # General
164
+ clipboard: bool = opts.CLIPBOARD,
165
+ log_level: str = opts.LOG_LEVEL,
166
+ log_file: str | None = opts.LOG_FILE,
167
+ quiet: bool = opts.QUIET,
168
+ ) -> None:
169
+ """Wyoming ASR Client for streaming microphone audio to a transcription server.
170
+
171
+ Usage:
172
+ - Run in foreground: agent-cli transcribe --device-index 1
173
+ - Run in background: agent-cli transcribe --device-index 1 &
174
+ - Check status: agent-cli transcribe --status
175
+ - Stop background process: agent-cli transcribe --stop
176
+ """
177
+ setup_logging(log_level, log_file, quiet=quiet)
178
+ console = Console() if not quiet else None
179
+ process_name = "transcribe"
180
+
181
+ if stop:
182
+ if process_manager.kill_process(process_name):
183
+ print_status_message(console, "✅ Transcribe stopped.")
184
+ else:
185
+ print_status_message(console, "⚠️ No transcribe is running.", style="yellow")
186
+ return
187
+
188
+ if status:
189
+ if process_manager.is_process_running(process_name):
190
+ pid = process_manager.read_pid_file(process_name)
191
+ print_status_message(console, f"✅ Transcribe is running (PID: {pid}).")
192
+ else:
193
+ print_status_message(console, "⚠️ Transcribe is not running.", style="yellow")
194
+ return
195
+
196
+ console = Console() if not quiet else None
197
+
198
+ with asr.pyaudio_context() as p:
199
+ if list_devices:
200
+ asr.list_input_devices(p, console)
201
+ return
202
+ device_index, device_name = asr.input_device(p, device_name, device_index)
203
+ print_device_index(console, device_index, device_name)
204
+
205
+ # Use context manager for PID file management
206
+ with process_manager.pid_file_context(process_name), suppress(KeyboardInterrupt):
207
+ asyncio.run(
208
+ async_main(
209
+ device_index=device_index,
210
+ asr_server_ip=asr_server_ip,
211
+ asr_server_port=asr_server_port,
212
+ clipboard=clipboard,
213
+ quiet=quiet,
214
+ model=model,
215
+ ollama_host=ollama_host,
216
+ llm=llm,
217
+ console=console,
218
+ p=p,
219
+ ),
220
+ )