feat(examples): add 540-livekit-voice-agent-python by github-actions[bot] · Pull Request #231 · deepgram/examples

github-actions · 2026-04-14T19:50:13Z

Closes #225

Auto-generated from issue suggestion by the engineering pipeline.

Action: New
Runtime: python
Example: 540-livekit-voice-agent-python

Build log

Build log: example 540-livekit-voice-agent-python

⚙️ Container started
ℹ️ Bootstrapping container tools...
ℹ️ Starting agent loop for example 540-livekit-voice-agent-python (max turns: 75)
ℹ️ Turn 1/75 — turn=1 phases=[] tests=not yet run history_len=0

`run_command`: `echo "DEEPGRAM_API_KEY is set: ${DEEPGRAM_API_KEY:+yes}" && echo "LIVEKIT_API_KEY is set: ${LIVEKIT_API_KEY:+yes}" && echo "LIVEKIT_API_SECRET is set: ${LIVEKIT_API_SECRET:+yes}" && echo "LIVEKIT_URL is set: ${LIVEKIT_URL:+yes}"`

exit: 0

DEEPGRAM_API_KEY is set: yes
LIVEKIT_API_KEY is set: yes
LIVEKIT_API_SECRET is set: 
LIVEKIT_URL is set: yes

ℹ️ Turn 2/75 — turn=2 phases=['api_verified'] tests=not yet run history_len=1

`run_command`: `deepgram --help 2>/dev/null || echo "deepgram CLI not available"`

exit: 0

deepgram CLI not available

ℹ️ Turn 3/75 — turn=3 phases=['api_verified'] tests=not yet run history_len=2

`run_command`: `context7 search "livekit agents python deepgram voice"`

exit: 127
stderr:

bash: line 1: context7: command not found

ℹ️ Turn 4/75 — turn=4 phases=['api_verified'] tests=not yet run history_len=3
ℹ️ Turn 5/75 — turn=5 phases=['api_verified'] tests=not yet run history_len=3

`run_command`: `pip install livekit-agents --dry-run 2>&1 | head -20`

exit: 0

Collecting livekit-agents
  Downloading livekit_agents-1.5.2-py3-none-any.whl.metadata (8.5 kB)
Collecting aiofiles>=24 (from livekit-agents)
  Downloading aiofiles-25.1.0-py3-none-any.whl.metadata (6.3 kB)
Collecting aiohttp~=3.10 (from livekit-agents)
  Downloading aiohttp-3.13.5-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (8.1 kB)
Collecting av>=14.0.0 (from livekit-agents)
  Downloading av-17.0.0-cp311-abi3-manylinux_2_28_x86_64.whl.metadata (4.6 kB)
Collecting certifi>=2025.6.15 (from livekit-agents)
  Downloading certifi-2026.2.25-py3-none-any.whl.metadata (2.5 kB)
Collecting click~=8.1 (from livekit-agents)
  Downloading click-8.3.2-py3-none-any.whl.metadata (2.6 kB)
Collecting colorama>=0.4.6 (from livekit-agents)
  Downloading colorama-0.4.6-py2.py3-none-any.whl.metadata (17 kB)
Collecting docstring-parser>=0.16 (from livekit-agents)
  Downloading docstring_parser-0.18.0-py3-none-any.whl.metadata (3.5 kB)
Collecting eval-type-backport (from livekit-agents)
  Downloading eval_type_backport-0.3.1-py3-none-any.whl.metadata (2.4 kB)
Collecting livekit-api<2,>=1.0.7 (from livekit-agents)
  Downloading livekit_api-1.1.0-py3-none-any.whl.metadata (1.5 kB)

ℹ️ Turn 6/75 — turn=6 phases=['api_verified'] tests=not yet run history_len=4

`run_command`: `pip search livekit-plugins-deepgram 2>&1 || pip index versions livekit-plugins-deepgram 2>&1 || pip install livekit-plugins-deepgram==9999 2>&1 | head -20`

exit: 0

ERROR: XMLRPC request failed [code: -32500]
RuntimeError: PyPI no longer supports 'pip search' (or XML-RPC search). Please use https://pypi.org/search (via a browser) instead. See https://warehouse.pypa.io/api-reference/xml-rpc.html#deprecated-methods for more information.
WARNING: pip index is currently an experimental command. It may be removed/changed in a future release without prior warning.
livekit-plugins-deepgram (1.5.2)
Available versions: 1.5.2, 1.5.1, 1.5.0, 1.4.6, 1.4.5, 1.4.4, 1.4.3, 1.4.2, 1.4.1, 1.3.12, 1.3.11, 1.3.10, 1.3.9, 1.3.8, 1.3.7, 1.3.6, 1.3.5, 1.3.4, 1.3.3, 1.3.2, 1.3.1, 1.2.18, 1.2.17, 1.2.16, 1.2.15, 1.2.14, 1.2.13, 1.2.12, 1.2.11, 1.2.9, 1.2.8, 1.2.7, 1.2.6, 1.2.5, 1.2.4, 1.2.3, 1.2.2, 1.2.1, 1.2.0, 1.1.7, 1.1.6, 1.1.5, 1.1.4, 1.1.3, 1.1.2, 1.1.1, 1.1.0, 1.0.23, 1.0.22, 1.0.21, 1.0.20, 1.0.19, 1.0.18, 1.0.17, 1.0.16, 1.0.15, 1.0.14, 1.0.13, 1.0.12, 1.0.11, 1.0.10, 1.0.2, 1.0.1, 1.0.0, 0.7.4, 0.7.3, 0.7.2, 0.7.1, 0.7.0, 0.6.20, 0.6.19, 0.6.18, 0.6.17, 0.6.16, 0.6.15, 0.6.14, 0.6.13, 0.6.11, 0.6.10, 0.6.9, 0.6.8, 0.6.7, 0.6.6, 0.6.5, 0.6.4, 0.6.3, 0.6.2, 0.6.1, 0.6.0, 0.5.1, 0.5.0, 0.4.0, 0.3.0, 0.2.0, 0.1.4, 0.1.3, 0.1.2, 0.1.1, 0.1.0, 0.0.5, 0.0.4, 0.0.3, 0.0.2, 0.0.1

[notice] A new release of pip is available: 25.0.1 -> 26.0.1
[notice] To update, run: pip install --upgrade pip

ℹ️ Turn 7/75 — turn=7 phases=['api_verified'] tests=not yet run history_len=5

`run_command`: `cat /workspace/../030-livekit-agents-python/README.md 2>/dev/null || echo "Looking for existing example..."`

exit: 0

Looking for existing example...

ℹ️ Turn 8/75 — turn=8 phases=['api_verified'] tests=not yet run history_len=6

`run_command`: `ls -la /workspace/../ 2>/dev/null | head -20 || echo "can't list parent"`

exit: 0

total 72
drwxr-xr-x   1 root root 4096 Apr 14 19:42 .
drwxr-xr-x   1 root root 4096 Apr 14 19:42 ..
-rwxr-xr-x   1 root root    0 Apr 14 19:42 .dockerenv
lrwxrwxrwx   1 root root    7 Apr  6 00:00 bin -> usr/bin
drwxr-xr-x   2 root root 4096 Jan  2 12:40 boot
drwxr-xr-x   5 root root  340 Apr 14 19:42 dev
drwxr-xr-x   1 root root 4096 Apr 14 19:42 etc
drwxr-xr-x   2 root root 4096 Jan  2 12:40 home
lrwxrwxrwx   1 root root    7 Apr  6 00:00 lib -> usr/lib
lrwxrwxrwx   1 root root    9 Apr  6 00:00 lib64 -> usr/lib64
drwxr-xr-x   2 root root 4096 Apr  6 00:00 media
drwxr-xr-x   2 root root 4096 Apr  6 00:00 mnt
drwxr-xr-x   2 root root 4096 Apr  6 00:00 opt
dr-xr-xr-x 229 root root    0 Apr 14 19:42 proc
drwx------   1 root root 4096 Apr 14 19:42 root
drwxr-xr-x   1 root root 4096 Apr  7 02:42 run
lrwxrwxrwx   1 root root    8 Apr  6 00:00 sbin -> usr/sbin
drwxr-xr-x   2 root root 4096 Apr  6 00:00 srv
dr-xr-xr-x  12 root root    0 Apr 14 19:42 sys

ℹ️ Turn 9/75 — turn=9 phases=['api_verified'] tests=not yet run history_len=7

`run_command`: `pip install livekit-agents livekit-plugins-deepgram livekit-plugins-silero --quiet 2>&1 | tail -5`

exit: 0

WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning.

[notice] A new release of pip is available: 25.0.1 -> 26.0.1
[notice] To update, run: pip install --upgrade pip

ℹ️ Turn 10/75 — turn=10 phases=['api_verified'] tests=not yet run history_len=8

`run_command`: `python3 -c "import livekit.agents; print(livekit.agents.version)"`

exit: 0

1.5.2

ℹ️ Turn 11/75 — turn=11 phases=['api_verified'] tests=not yet run history_len=9

`run_command`: `python3 -c "from livekit.plugins import deepgram; print(dir(deepgram))"`

exit: 0

['DeepgramPlugin', 'NOT_IN_ALL', 'Plugin', 'STT', 'STTv2', 'SpeechStream', 'SpeechStreamv2', 'TTS', '__all__', '__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__pdoc__', '__spec__', '__version__', '_module', '_utils', 'log', 'logger', 'models', 'n', 'stt', 'stt_v2', 'tts', 'version']

ℹ️ Turn 12/75 — turn=12 phases=['api_verified'] tests=not yet run history_len=10

`run_command`: `python3 -c "from livekit.plugins.deepgram import STT, TTS; help(STT)" 2>&1 | head -80`

exit: 0

Help on class STT in module livekit.plugins.deepgram.stt:

class STT(livekit.agents.stt.stt.STT)
 |  STT(*, model: 'DeepgramModels | str' = 'nova-3', language: 'DeepgramLanguages | str' = 'en-US', detect_language: 'bool' = False, interim_results: 'bool' = True, punctuate: 'bool' = True, smart_format: 'bool' = False, sample_rate: 'int' = 16000, no_delay: 'bool' = True, endpointing_ms: 'int' = 25, enable_diarization: 'bool' = False, filler_words: 'bool' = True, keywords: 'NotGivenOr[list[tuple[str, float]]]' = NOT_GIVEN, keyterm: 'NotGivenOr[str | list[str]]' = NOT_GIVEN, tags: 'NotGivenOr[list[str]]' = NOT_GIVEN, profanity_filter: 'bool' = False, api_key: 'NotGivenOr[str]' = NOT_GIVEN, http_session: 'aiohttp.ClientSession | None' = None, base_url: 'str' = 'https://api.deepgram.com/v1/listen', numerals: 'bool' = False, mip_opt_out: 'bool' = False, vad_events: 'bool' = True, keyterms: 'NotGivenOr[list[str]]' = NOT_GIVEN) -> 'None'
 |
 |  Method resolution order:
 |      STT
 |      livekit.agents.stt.stt.STT
 |      abc.ABC
 |      livekit.rtc.event_emitter.EventEmitter
 |      typing.Generic
 |      builtins.object
 |
 |  Methods defined here:
 |
 |  __init__(self, *, model: 'DeepgramModels | str' = 'nova-3', language: 'DeepgramLanguages | str' = 'en-US', detect_language: 'bool' = False, interim_results: 'bool' = True, punctuate: 'bool' = True, smart_format: 'bool' = False, sample_rate: 'int' = 16000, no_delay: 'bool' = True, endpointing_ms: 'int' = 25, enable_diarization: 'bool' = False, filler_words: 'bool' = True, keywords: 'NotGivenOr[list[tuple[str, float]]]' = NOT_GIVEN, keyterm: 'NotGivenOr[str | list[str]]' = NOT_GIVEN, tags: 'NotGivenOr[list[str]]' = NOT_GIVEN, profanity_filter: 'bool' = False, api_key: 'NotGivenOr[str]' = NOT_GIVEN, http_session: 'aiohttp.ClientSession | None' = None, base_url: 'str' = 'https://api.deepgram.com/v1/listen', numerals: 'bool' = False, mip_opt_out: 'bool' = False, vad_events: 'bool' = True, keyterms: 'NotGivenOr[list[str]]' = NOT_GIVEN) -> 'None'
 |      Create a new instance of Deepgram STT.
 |
 |      Args:
 |          model: The Deepgram model to use for speech recognition. Defaults to "nova-3".
 |          language: The language code for recognition. Defaults to "en-US".
 |          detect_language: Whether to enable automatic language detection. Defaults to False.
 |          interim_results: Whether to return interim (non-final) transcription results. Defaults to True.
 |          punctuate: Whether to add punctuations to the transcription. Defaults to True. Turn detector will work better with punctuations.
 |          smart_format: Whether to apply smart formatting to numbers, dates, etc. Defaults to False.
 |          sample_rate: The sample rate of the audio in Hz. Defaults to 16000.
 |          no_delay: When smart_format is used, ensures it does not wait for sequence to be complete before returning results. Defaults to True.
 |          endpointing_ms: Time in milliseconds of silence to consider end of speech. Set to 0 to disable. Defaults to 25.
 |          filler_words: Whether to include filler words (um, uh, etc.) in transcription. Defaults to True.
 |          keywords: List of tuples containing keywords and their boost values for improved recognition.
 |                   Each tuple should be (keyword: str, boost: float). Defaults to None.
 |                   `keywords` does not work with Nova-3 models. Use `keyterm` instead.
 |          keyterm: str or list of str of key terms to improve recognition accuracy. Defaults to None.
 |                   `keyterm` is only supported by Nova-3 models.
 |          tags: List of tags to add to the requests for usage reporting. Defaults to NOT_GIVEN.
 |          profanity_filter: Whether to filter profanity from the transcription. Defaults to False.
 |          api_key: Your Deepgram API key. If not provided, will look for DEEPGRAM_API_KEY environment variable.
 |          http_session: Optional aiohttp ClientSession to use for requ

ℹ️ Turn 13/75 — turn=13 phases=['api_verified'] tests=not yet run history_len=11

`run_command`: `python3 -c "from livekit.plugins.deepgram import TTS; help(TTS)" 2>&1 | head -80`

exit: 0

Help on class TTS in module livekit.plugins.deepgram.tts:

class TTS(livekit.agents.tts.tts.TTS)
 |  TTS(*, model: 'str' = 'aura-2-andromeda-en', encoding: 'str' = 'linear16', sample_rate: 'int' = 24000, api_key: 'str | None' = None, base_url: 'str' = 'https://api.deepgram.com/v1/speak', word_tokenizer: 'NotGivenOr[tokenize.WordTokenizer]' = NOT_GIVEN, http_session: 'aiohttp.ClientSession | None' = None, mip_opt_out: 'bool' = False) -> 'None'
 |
 |  Method resolution order:
 |      TTS
 |      livekit.agents.tts.tts.TTS
 |      abc.ABC
 |      livekit.rtc.event_emitter.EventEmitter
 |      typing.Generic
 |      builtins.object
 |
 |  Methods defined here:
 |
 |  __init__(self, *, model: 'str' = 'aura-2-andromeda-en', encoding: 'str' = 'linear16', sample_rate: 'int' = 24000, api_key: 'str | None' = None, base_url: 'str' = 'https://api.deepgram.com/v1/speak', word_tokenizer: 'NotGivenOr[tokenize.WordTokenizer]' = NOT_GIVEN, http_session: 'aiohttp.ClientSession | None' = None, mip_opt_out: 'bool' = False) -> 'None'
 |      Create a new instance of Deepgram TTS.
 |
 |      Args:
 |          model (str): TTS model to use. Defaults to "aura-2-andromeda-en".
 |          encoding (str): Audio encoding to use. Defaults to "linear16".
 |          sample_rate (int): Sample rate of audio. Defaults to 24000.
 |          api_key (str): Deepgram API key. If not provided, will look for DEEPGRAM_API_KEY in environment.
 |          base_url (str): Base URL for Deepgram TTS API. Defaults to "https://api.deepgram.com/v1/speak"
 |          word_tokenizer (tokenize.WordTokenizer): Tokenizer for processing text. Defaults to basic WordTokenizer.
 |          http_session (aiohttp.ClientSession): Optional aiohttp session to use for requests.
 |
 |  async aclose(self) -> 'None'
 |
 |  prewarm(self) -> 'None'
 |      Pre-warm connection to the TTS service
 |
 |  stream(self, *, conn_options: 'APIConnectOptions' = APIConnectOptions(max_retry=3, retry_interval=2.0, timeout=10.0)) -> 'SynthesizeStream'
 |
 |  synthesize(self, text: 'str', *, conn_options: 'APIConnectOptions' = APIConnectOptions(max_retry=3, retry_interval=2.0, timeout=10.0)) -> 'ChunkedStream'
 |
 |  update_options(self, *, model: 'NotGivenOr[str]' = NOT_GIVEN) -> 'None'
 |      Args:
 |          model (str): TTS model to use.
 |
 |  ----------------------------------------------------------------------
 |  Readonly properties defined here:
 |
 |  model
 |      Get the model name/identifier for this TTS instance.
 |
 |      Returns:
 |          The model name if available, "unknown" otherwise.
 |
 |      Note:
 |          Plugins should override this property to provide their model information.
 |
 |  provider
 |      Get the provider name/identifier for this TTS instance.
 |
 |      Returns:
 |          The provider name if available, "unknown" otherwise.
 |
 |      Note:
 |          Plugins should override this property to provide their provider information.
 |
 |  ----------------------------------------------------------------------
 |  Data and other attributes defined here:
 |
 |  __abstractmethods__ = frozenset()
 |
 |  __parameters__ = ()
 |
 |  ----------------------------------------------------------------------
 |  Methods inherited from livekit.agents.tts.tts.TTS:
 |
 |  async __aenter__(self) -> 'TTS'
 |
 |  async __aexit__(self, exc_type: 'type[BaseException] | None', exc: 'BaseException | None', exc_tb: 'TracebackType | None') -> 'None'
 |
 |  ----------------------------------------------------------------------
 |  Readonly properties inherited from livekit.agents.tts.tts.TTS:
 |
 |  capabilities
 |

ℹ️ Turn 14/75 — turn=14 phases=['api_verified'] tests=not yet run history_len=12

`run_command`: `python3 -c "from livekit.agents import AgentSession, Agent; help(AgentSession)" 2>&1 | head -100`

exit: 0

Help on class AgentSession in module livekit.agents.voice.agent_session:

class AgentSession(livekit.rtc.event_emitter.EventEmitter, typing.Generic)
 |  AgentSession(*, stt: 'NotGivenOr[stt.STT | STTModels | str]' = NOT_GIVEN, vad: 'NotGivenOr[vad.VAD]' = NOT_GIVEN, llm: 'NotGivenOr[llm.LLM | llm.RealtimeModel | LLMModels | str]' = NOT_GIVEN, tts: 'NotGivenOr[tts.TTS | TTSModels | str]' = NOT_GIVEN, turn_handling: 'NotGivenOr[TurnHandlingOptions]' = NOT_GIVEN, tools: 'NotGivenOr[list[llm.Tool | llm.Toolset]]' = NOT_GIVEN, mcp_servers: 'NotGivenOr[list[mcp.MCPServer]]' = NOT_GIVEN, max_tool_steps: 'int' = 3, use_tts_aligned_transcript: 'NotGivenOr[bool]' = NOT_GIVEN, tts_text_transforms: 'NotGivenOr[Sequence[TextTransforms] | None]' = NOT_GIVEN, min_consecutive_speech_delay: 'float' = 0.0, userdata: 'NotGivenOr[Userdata_T]' = NOT_GIVEN, video_sampler: 'NotGivenOr[_VideoSampler | None]' = NOT_GIVEN, preemptive_generation: 'bool' = True, aec_warmup_duration: 'float | None' = 3.0, ivr_detection: 'bool' = False, user_away_timeout: 'float | None' = 15.0, conn_options: 'NotGivenOr[SessionConnectOptions]' = NOT_GIVEN, loop: 'asyncio.AbstractEventLoop | None' = None, min_endpointing_delay: 'NotGivenOr[float]' = NOT_GIVEN, max_endpointing_delay: 'NotGivenOr[float]' = NOT_GIVEN, false_interruption_timeout: 'NotGivenOr[float | None]' = NOT_GIVEN, turn_detection: 'NotGivenOr[TurnDetectionMode]' = NOT_GIVEN, discard_audio_if_uninterruptible: 'NotGivenOr[bool]' = NOT_GIVEN, min_interruption_duration: 'NotGivenOr[float]' = NOT_GIVEN, min_interruption_words: 'NotGivenOr[int]' = NOT_GIVEN, allow_interruptions: 'NotGivenOr[bool]' = NOT_GIVEN, resume_false_interruption: 'NotGivenOr[bool]' = NOT_GIVEN, agent_false_interruption_timeout: 'NotGivenOr[float | None]' = NOT_GIVEN) -> 'None'
 |
 |  Method resolution order:
 |      AgentSession
 |      livekit.rtc.event_emitter.EventEmitter
 |      typing.Generic
 |      builtins.object
 |
 |  Methods defined here:
 |
 |  async __aenter__(self) -> 'AgentSession'
 |
 |  async __aexit__(self, exc_type: 'type[BaseException] | None', exc: 'BaseException | None', exc_tb: 'TracebackType | None') -> 'None'
 |
 |  __init__(self, *, stt: 'NotGivenOr[stt.STT | STTModels | str]' = NOT_GIVEN, vad: 'NotGivenOr[vad.VAD]' = NOT_GIVEN, llm: 'NotGivenOr[llm.LLM | llm.RealtimeModel | LLMModels | str]' = NOT_GIVEN, tts: 'NotGivenOr[tts.TTS | TTSModels | str]' = NOT_GIVEN, turn_handling: 'NotGivenOr[TurnHandlingOptions]' = NOT_GIVEN, tools: 'NotGivenOr[list[llm.Tool | llm.Toolset]]' = NOT_GIVEN, mcp_servers: 'NotGivenOr[list[mcp.MCPServer]]' = NOT_GIVEN, max_tool_steps: 'int' = 3, use_tts_aligned_transcript: 'NotGivenOr[bool]' = NOT_GIVEN, tts_text_transforms: 'NotGivenOr[Sequence[TextTransforms] | None]' = NOT_GIVEN, min_consecutive_speech_delay: 'float' = 0.0, userdata: 'NotGivenOr[Userdata_T]' = NOT_GIVEN, video_sampler: 'NotGivenOr[_VideoSampler | None]' = NOT_GIVEN, preemptive_generation: 'bool' = True, aec_warmup_duration: 'float | None' = 3.0, ivr_detection: 'bool' = False, user_away_timeout: 'float | None' = 15.0, conn_options: 'NotGivenOr[SessionConnectOptions]' = NOT_GIVEN, loop: 'asyncio.AbstractEventLoop | None' = None, min_endpointing_delay: 'NotGivenOr[float]' = NOT_GIVEN, max_endpointing_delay: 'NotGivenOr[float]' = NOT_GIVEN, false_interruption_timeout: 'NotGivenOr[float | None]' = NOT_GIVEN, turn_detection: 'NotGivenOr[TurnDetectionMode]' = NOT_GIVEN, discard_audio_if_uninterruptible: 'NotGivenOr[bool]' = NOT_GIVEN, min_interruption_duration: 'NotGivenOr[float]' = NOT_GIVEN, min_interruption_words: 'NotGivenOr[int]' = NOT_GIVEN, allow_interruptions: 'NotGivenOr[bool]' = NOT_GIVEN, resume_false_interruption: 'NotGivenOr[bool]' = NOT_GIVEN, agent_false_interruption_timeout: 'NotGivenOr[float | None]' = NOT_GIVEN) -> 'None'
 |      `AgentSession` is the LiveKit Agents runtime that glues together
 |      media streams, speech/LLM components, and tool orchestration into a
 |      single real-time voice agent.
 |

ℹ️ Turn 15/75 — turn=15 phases=['api_verified'] tests=not yet run history_len=13

`run_command`: `python3 -c "from livekit.agents import Agent; help(Agent)" 2>&1 | head -60`

exit: 0

Help on class Agent in module livekit.agents.voice.agent:

class Agent(builtins.object)
 |  Agent(*, instructions: 'str | Instructions', id: 'str | None' = None, chat_ctx: 'NotGivenOr[llm.ChatContext | None]' = NOT_GIVEN, tools: 'list[llm.Tool | llm.Toolset] | None' = None, stt: 'NotGivenOr[stt.STT | STTModels | str | None]' = NOT_GIVEN, vad: 'NotGivenOr[vad.VAD | None]' = NOT_GIVEN, turn_handling: 'NotGivenOr[TurnHandlingOptions]' = NOT_GIVEN, llm: 'NotGivenOr[llm.LLM | llm.RealtimeModel | LLMModels | str | None]' = NOT_GIVEN, tts: 'NotGivenOr[tts.TTS | TTSModels | str | None]' = NOT_GIVEN, mcp_servers: 'NotGivenOr[list[mcp.MCPServer] | None]' = NOT_GIVEN, min_consecutive_speech_delay: 'NotGivenOr[float]' = NOT_GIVEN, use_tts_aligned_transcript: 'NotGivenOr[bool]' = NOT_GIVEN, turn_detection: 'NotGivenOr[TurnDetectionMode | None]' = NOT_GIVEN, min_endpointing_delay: 'NotGivenOr[float]' = NOT_GIVEN, max_endpointing_delay: 'NotGivenOr[float]' = NOT_GIVEN, allow_interruptions: 'NotGivenOr[bool]' = NOT_GIVEN) -> 'None'
 |
 |  Methods defined here:
 |
 |  __init__(self, *, instructions: 'str | Instructions', id: 'str | None' = None, chat_ctx: 'NotGivenOr[llm.ChatContext | None]' = NOT_GIVEN, tools: 'list[llm.Tool | llm.Toolset] | None' = None, stt: 'NotGivenOr[stt.STT | STTModels | str | None]' = NOT_GIVEN, vad: 'NotGivenOr[vad.VAD | None]' = NOT_GIVEN, turn_handling: 'NotGivenOr[TurnHandlingOptions]' = NOT_GIVEN, llm: 'NotGivenOr[llm.LLM | llm.RealtimeModel | LLMModels | str | None]' = NOT_GIVEN, tts: 'NotGivenOr[tts.TTS | TTSModels | str | None]' = NOT_GIVEN, mcp_servers: 'NotGivenOr[list[mcp.MCPServer] | None]' = NOT_GIVEN, min_consecutive_speech_delay: 'NotGivenOr[float]' = NOT_GIVEN, use_tts_aligned_transcript: 'NotGivenOr[bool]' = NOT_GIVEN, turn_detection: 'NotGivenOr[TurnDetectionMode | None]' = NOT_GIVEN, min_endpointing_delay: 'NotGivenOr[float]' = NOT_GIVEN, max_endpointing_delay: 'NotGivenOr[float]' = NOT_GIVEN, allow_interruptions: 'NotGivenOr[bool]' = NOT_GIVEN) -> 'None'
 |      Initialize self.  See help(type(self)) for accurate signature.
 |
 |  llm_node(self, chat_ctx: 'llm.ChatContext', tools: 'list[llm.Tool]', model_settings: 'ModelSettings') -> 'AsyncIterable[llm.ChatChunk | str | FlushSentinel] | Coroutine[Any, Any, AsyncIterable[llm.ChatChunk | str | FlushSentinel]] | Coroutine[Any, Any, str] | Coroutine[Any, Any, llm.ChatChunk] | Coroutine[Any, Any, None]'
 |      A node in the processing pipeline that processes text generation with an LLM.
 |
 |      By default, this node uses the agent's LLM to process the provided context. It may yield
 |      plain text (as `str`) for straightforward text generation, or `llm.ChatChunk` objects that
 |      can include text and optional tool calls. `ChatChunk` is helpful for capturing more complex
 |      outputs such as function calls, usage statistics, or other metadata.
 |
 |      You can override this node to customize how the LLM is used or how tool invocations
 |      and responses are handled.
 |
 |      Args:
 |          chat_ctx (llm.ChatContext): The context for the LLM (the conversation history).
 |          tools (list[FunctionTool]): A list of callable tools that the LLM may invoke.
 |          model_settings (ModelSettings): Configuration and parameters for model execution.
 |
 |      Yields/Returns:
 |          str: Plain text output from the LLM.
 |          llm.ChatChunk: An object that can contain both text and optional tool calls.
 |
 |  async on_enter(self) -> 'None'
 |      Called when the task is entered
 |
 |  async on_exit(self) -> 'None'
 |      Called when the task is exited
 |
 |  async on_user_turn_completed(self, turn_ctx: 'llm.ChatContext', new_message: 'llm.ChatMessage') -> 'None'
 |      Called when the user has finished speaking, and the LLM is about to respond
 |
 |      This is a good opportunity to update the chat context or edit the new message before it is
 |      sent to the LLM.
 |
 |  realtime_audio_output_node(self, audio: 'Async

ℹ️ Turn 16/75 — turn=16 phases=['api_verified'] tests=not yet run history_len=14

`run_command`: `python3 -c "from livekit.agents import cli; help(cli)" 2>&1 | head -50`

exit: 0

Help on package livekit.agents.cli in livekit.agents:

NAME
    livekit.agents.cli

PACKAGE CONTENTS
    cli
    discover
    log
    proto
    readchar
    watcher

CLASSES
    builtins.object
        livekit.agents.cli.cli.AgentsConsole

    class AgentsConsole(builtins.object)
     |  AgentsConsole() -> 'None'
     |
     |  Methods defined here:
     |
     |  __init__(self) -> 'None'
     |      Initialize self.  See help(type(self)) for accurate signature.
     |
     |  acquire_io(self, *, loop: 'asyncio.AbstractEventLoop', session: 'AgentSession') -> 'None'
     |
     |  print(self, child: 'RenderableType', *, tag: 'str' = '', tag_style: 'Style | None' = None) -> 'None'
     |
     |  set_microphone_enabled(self, enable: 'bool', *, device: 'int | str | None' = None) -> 'None'
     |
     |  set_speaker_enabled(self, enable: 'bool', *, device: 'int | str | None' = None) -> 'None'
     |
     |  wait_for_io_acquisition(self) -> 'None'
     |
     |  ----------------------------------------------------------------------
     |  Class methods defined here:
     |
     |  get_instance() -> 'AgentsConsole'
     |
     |  ----------------------------------------------------------------------
     |  Readonly properties defined here:
     |
     |  input_name
     |
     |  io_acquired
     |
     |  io_context
     |
     |  io_loop

ℹ️ Turn 17/75 — turn=17 phases=['api_verified'] tests=not yet run history_len=15

`run_command`: `python3 -c "from livekit.agents import cli; print(dir(cli))"`

exit: 0

['AgentsConsole', 'NOT_IN_ALL', '__all__', '__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__pdoc__', '__spec__', '_module', 'cli', 'log', 'n', 'proto', 'readchar', 'run_app']

ℹ️ Turn 18/75 — turn=18 phases=['api_verified'] tests=not yet run history_len=16

`run_command`: `python3 -c "from livekit.agents.cli import run_app; help(run_app)" 2>&1 | head -100`

exit: 0

Help on function run_app in module livekit.agents.cli.cli:

run_app(server: 'AgentServer | WorkerOptions') -> 'None'

ℹ️ Turn 19/75 — turn=19 phases=['api_verified'] tests=not yet run history_len=17

`run_command`: `python3 -c "from livekit.agents import AgentServer; help(AgentServer)" 2>&1 | head -50`

exit: 0

Help on class AgentServer in module livekit.agents.worker:

class AgentServer(livekit.rtc.event_emitter.EventEmitter)
 |  AgentServer(*, job_executor_type: 'JobExecutorType' = <JobExecutorType.PROCESS: 'process'>, load_threshold: 'float | ServerEnvOption[float]' = ServerEnvOption(dev_default=inf, prod_default=0.7), job_memory_warn_mb: 'float' = 500, job_memory_limit_mb: 'float' = 0, drain_timeout: 'int' = 1800, num_idle_processes: 'int | ServerEnvOption[int]' = ServerEnvOption(dev_default=0, prod_default=2), shutdown_process_timeout: 'float' = 10.0, session_end_timeout: 'float' = 300.0, initialize_process_timeout: 'float' = 10.0, permissions: 'WorkerPermissions' = WorkerPermissions(can_publish=True, can_subscribe=True, can_publish_data=True, can_update_metadata=True, can_publish_sources=[], hidden=False), max_retry: 'int' = 16, ws_url: 'str | None' = None, api_key: 'str | None' = None, api_secret: 'str | None' = None, host: 'str' = '', port: 'int | ServerEnvOption[int]' = ServerEnvOption(dev_default=0, prod_default=8081), http_proxy: 'NotGivenOr[str | None]' = NOT_GIVEN, multiprocessing_context: "Literal['spawn', 'forkserver']" = 'forkserver', setup_fnc: 'Callable[[JobProcess], Any] | None' = None, load_fnc: 'Callable[[AgentServer], float] | Callable[[], float] | None' = None, prometheus_port: 'int | None' = None, prometheus_multiproc_dir: 'str | None' = None, log_level: 'str | ServerEnvOption[str]' = ServerEnvOption(dev_default='DEBUG', prod_default='INFO')) -> 'None'
 |
 |  Method resolution order:
 |      AgentServer
 |      livekit.rtc.event_emitter.EventEmitter
 |      typing.Generic
 |      builtins.object
 |
 |  Methods defined here:
 |
 |  __init__(self, *, job_executor_type: 'JobExecutorType' = <JobExecutorType.PROCESS: 'process'>, load_threshold: 'float | ServerEnvOption[float]' = ServerEnvOption(dev_default=inf, prod_default=0.7), job_memory_warn_mb: 'float' = 500, job_memory_limit_mb: 'float' = 0, drain_timeout: 'int' = 1800, num_idle_processes: 'int | ServerEnvOption[int]' = ServerEnvOption(dev_default=0, prod_default=2), shutdown_process_timeout: 'float' = 10.0, session_end_timeout: 'float' = 300.0, initialize_process_timeout: 'float' = 10.0, permissions: 'WorkerPermissions' = WorkerPermissions(can_publish=True, can_subscribe=True, can_publish_data=True, can_update_metadata=True, can_publish_sources=[], hidden=False), max_retry: 'int' = 16, ws_url: 'str | None' = None, api_key: 'str | None' = None, api_secret: 'str | None' = None, host: 'str' = '', port: 'int | ServerEnvOption[int]' = ServerEnvOption(dev_default=0, prod_default=8081), http_proxy: 'NotGivenOr[str | None]' = NOT_GIVEN, multiprocessing_context: "Literal['spawn', 'forkserver']" = 'forkserver', setup_fnc: 'Callable[[JobProcess], Any] | None' = None, load_fnc: 'Callable[[AgentServer], float] | Callable[[], float] | None' = None, prometheus_port: 'int | None' = None, prometheus_multiproc_dir: 'str | None' = None, log_level: 'str | ServerEnvOption[str]' = ServerEnvOption(dev_default='DEBUG', prod_default='INFO')) -> 'None'
 |      Initialize a new instance of EventEmitter.
 |
 |  async aclose(self) -> 'None'
 |
 |  async drain(self, timeout: 'NotGivenOr[int | None]' = NOT_GIVEN) -> 'None'
 |      When timeout isn't None, it will raise asyncio.TimeoutError if the processes didn't finish in time.
 |
 |  rtc_session(self, func: 'Callable[[JobContext], Awaitable[None]] | None' = None, *, agent_name: 'str' = '', type: 'ServerType' = <ServerType.ROOM: 0>, on_request: 'Callable[[JobRequest], Any] | None' = None, on_session_end: 'Callable[[JobContext], Any] | None' = None) -> 'Callable[[JobContext], Awaitable[None]] | Callable[[Callable[[JobContext], Awaitable[None]]], Callable[[JobContext], Awaitable[None]]]'
 |      Decorator or direct registrar for the RTC session entrypoint.
 |
 |      Usage:
 |          @server.rtc_session(agent_name="survey_agent")
 |          async def my_agent(job_ctx: JobContext): ...
 |
 |          server.rtc_session(my_agent, agent_na

ℹ️ Turn 20/75 — turn=20 phases=['api_verified'] tests=not yet run history_len=18

`run_command`: `pip install livekit-plugins-openai --quiet 2>&1 | tail -3`

exit: 0


[notice] A new release of pip is available: 25.0.1 -> 26.0.1
[notice] To update, run: pip install --upgrade pip

ℹ️ Turn 21/75 — turn=21 phases=['api_verified'] tests=not yet run history_len=19

`run_command`: `python3 -c "from livekit.plugins import openai; print(dir(openai))"`

exit: 0

['EmbeddingData', 'LLM', 'LLMStream', 'NOT_IN_ALL', 'OpenAIPlugin', 'OpenRouterProviderPreferences', 'OpenRouterWebPlugin', 'Plugin', 'STT', 'STTModels', 'TTS', 'TTSModels', 'TTSVoices', '__all__', '__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__pdoc__', '__spec__', '__version__', '_module', 'create_embeddings', 'embeddings', 'llm', 'log', 'logger', 'models', 'n', 'realtime', 'responses', 'stt', 'tools', 'tts', 'utils', 'version']

ℹ️ Turn 22/75 — turn=22 phases=['api_verified'] tests=not yet run history_len=20

`run_command`: `pip index versions livekit-plugins-silero 2>&1 | head -5`

exit: 0

WARNING: pip index is currently an experimental command. It may be removed/changed in a future release without prior warning.
livekit-plugins-silero (1.5.2)
Available versions: 1.5.2, 1.5.1, 1.5.0, 1.4.6, 1.4.5, 1.4.4, 1.4.3, 1.4.2, 1.4.1, 1.3.12, 1.3.11, 1.3.10, 1.3.9, 1.3.8, 1.3.7, 1.3.6, 1.3.5, 1.3.4, 1.3.3, 1.3.2, 1.3.1, 1.2.18, 1.2.17, 1.2.16, 1.2.15, 1.2.14, 1.2.13, 1.2.12, 1.2.11, 1.2.9, 1.2.8, 1.2.7, 1.2.6, 1.2.5, 1.2.4, 1.2.3, 1.2.2, 1.2.1, 1.2.0, 1.1.7, 1.1.6, 1.1.5, 1.1.4, 1.1.3, 1.1.2, 1.1.1, 1.1.0, 1.0.23, 1.0.22, 1.0.21, 1.0.20, 1.0.19, 1.0.18, 1.0.17, 1.0.16, 1.0.15, 1.0.14, 1.0.13, 1.0.12, 1.0.11, 0.7.6, 0.7.5, 0.7.4, 0.7.3, 0.7.2, 0.7.1, 0.7.0, 0.6.4, 0.6.3, 0.6.2, 0.6.1, 0.6.0, 0.5.3, 0.5.2, 0.5.1, 0.5.0, 0.4.0, 0.3.0, 0.2.0, 0.1.1, 0.1.0, 0.0.8, 0.0.7
  INSTALLED: 1.5.2
  LATEST:    1.5.2

ℹ️ Turn 23/75 — turn=23 phases=['api_verified'] tests=not yet run history_len=21

`run_command`: `echo "OPENAI_API_KEY is set: ${OPENAI_API_KEY:+yes}"`

exit: 0

OPENAI_API_KEY is set:

ℹ️ Turn 24/75 — turn=24 phases=['api_verified'] tests=not yet run history_len=22

`run_command`: `pip search livekit-plugins 2>&1 || pip install livekit-plugins-anthropic==9999 2>&1 | head -15`

exit: 0

ERROR: XMLRPC request failed [code: -32500]
RuntimeError: PyPI no longer supports 'pip search' (or XML-RPC search). Please use https://pypi.org/search (via a browser) instead. See https://warehouse.pypa.io/api-reference/xml-rpc.html#deprecated-methods for more information.
ERROR: Could not find a version that satisfies the requirement livekit-plugins-anthropic==9999 (from versions: 0.0.1, 0.2.0, 0.2.1, 0.2.2, 0.2.3, 0.2.4, 0.2.5, 0.2.6, 0.2.7, 0.2.8, 0.2.9, 0.2.10, 0.2.11, 0.2.12, 0.2.13, 0.2.14, 1.0.0.dev4, 1.0.0.dev5, 1.0.0rc1, 1.0.0rc2, 1.0.0rc3, 1.0.0rc4, 1.0.0rc5, 1.0.0rc6, 1.0.0rc7, 1.0.0rc8, 1.0.0rc9, 1.0.0, 1.0.1, 1.0.2, 1.0.10, 1.0.11, 1.0.12, 1.0.13, 1.0.14, 1.0.15, 1.0.16, 1.0.17, 1.0.18, 1.0.19, 1.0.20, 1.0.21, 1.0.22, 1.0.23, 1.1.0, 1.1.1, 1.1.2, 1.1.3, 1.1.4, 1.1.5, 1.1.6, 1.1.7, 1.2.0, 1.2.1, 1.2.2, 1.2.3, 1.2.4, 1.2.5, 1.2.6, 1.2.7, 1.2.8, 1.2.9, 1.2.10, 1.2.11, 1.2.12, 1.2.13, 1.2.14, 1.2.15, 1.2.16, 1.2.17, 1.2.18, 1.3.0rc1, 1.3.0rc2, 1.3.1, 1.3.2, 1.3.3, 1.3.4, 1.3.5, 1.3.6, 1.3.7, 1.3.8, 1.3.9, 1.3.10, 1.3.11, 1.3.12, 1.4.0rc1, 1.4.0rc2, 1.4.0, 1.4.1, 1.4.2, 1.4.3, 1.4.4, 1.4.5, 1.4.6, 1.5.0rc1, 1.5.0rc2, 1.5.0, 1.5.1, 1.5.2)

[notice] A new release of pip is available: 25.0.1 -> 26.0.1
[notice] To update, run: pip install --upgrade pip
ERROR: No matching distribution found for livekit-plugins-anthropic==9999

ℹ️ Turn 25/75 — turn=25 phases=['api_verified'] tests=not yet run history_len=23

`run_command`: `echo "ANTHROPIC_API_KEY is set: ${ANTHROPIC_API_KEY:+yes}"`

exit: 0

ANTHROPIC_API_KEY is set:

ℹ️ Turn 26/75 — turn=26 phases=['api_verified'] tests=not yet run history_len=24

`run_command`: `pip install livekit-plugins-google==9999 2>&1 | head -5`

exit: 0

ERROR: Could not find a version that satisfies the requirement livekit-plugins-google==9999 (from versions: 0.0.2, 0.1.0, 0.1.1, 0.1.2, 0.2.0, 0.3.dev0, 0.3.0, 0.4.dev0, 0.4.dev1, 0.4.0, 0.5.dev0, 0.5.0, 0.5.1, 0.6.dev0, 0.6.0.dev1, 0.6.0.dev2, 0.6.0.dev7, 0.6.0, 0.6.1, 0.6.2, 0.6.3, 0.7.0, 0.7.1, 0.7.2, 0.7.3, 0.8.1, 0.9.0, 0.9.1, 0.10.0, 0.10.1, 0.10.2, 0.10.3, 0.10.4, 0.10.5, 0.10.6, 0.11.0, 0.11.1, 0.11.2, 0.11.3, 0.11.4, 0.11.5, 1.0.0.dev4, 1.0.0.dev5, 1.0.0rc1, 1.0.0rc2, 1.0.0rc3, 1.0.0rc4, 1.0.0rc5, 1.0.0rc6, 1.0.0rc7, 1.0.0rc8, 1.0.0rc9, 1.0.0, 1.0.1, 1.0.2, 1.0.10, 1.0.11, 1.0.12, 1.0.13, 1.0.14, 1.0.15, 1.0.16, 1.0.17, 1.0.18, 1.0.19, 1.0.20, 1.0.21, 1.0.22, 1.0.23, 1.1.0, 1.1.1, 1.1.2, 1.1.3, 1.1.4, 1.1.5, 1.1.6, 1.1.7, 1.2.0, 1.2.1, 1.2.2, 1.2.3, 1.2.4, 1.2.5, 1.2.6, 1.2.7, 1.2.8, 1.2.9, 1.2.11, 1.2.12, 1.2.13, 1.2.14, 1.2.15, 1.2.16, 1.2.17, 1.2.18, 1.3.0rc1, 1.3.0rc2, 1.3.1, 1.3.2, 1.3.3, 1.3.4, 1.3.5, 1.3.6, 1.3.7, 1.3.8, 1.3.9, 1.3.10, 1.3.11, 1.3.12, 1.4.0rc1, 1.4.0rc2, 1.4.1, 1.4.2, 1.4.3, 1.4.4, 1.4.5, 1.4.6, 1.5.0rc1, 1.5.0rc2, 1.5.0, 1.5.1, 1.5.2)

[notice] A new release of pip is available: 25.0.1 -> 26.0.1
[notice] To update, run: pip install --upgrade pip
ERROR: No matching distribution found for livekit-plugins-google==9999

ℹ️ Turn 27/75 — turn=27 phases=['api_verified'] tests=not yet run history_len=25

`run_command`: `echo "GOOGLE_API_KEY is set: ${GOOGLE_API_KEY:+yes}"`

exit: 0

GOOGLE_API_KEY is set:

ℹ️ Turn 28/75 — turn=28 phases=['api_verified'] tests=not yet run history_len=26

`run_command`: `echo "GROQ_API_KEY is set: ${GROQ_API_KEY:+yes}"`

exit: 0

GROQ_API_KEY is set:

ℹ️ Turn 29/75 — turn=29 phases=['api_verified'] tests=not yet run history_len=27

`run_command`: `env | grep -i "_API_KEY\|_API_SECRET\|_URL" | sed 's/=.*/=***/' | sort`

exit: 0

DEEPGRAM_API_KEY=***
LIVEKIT_API_KEY=***
LIVEKIT_URL=***

ℹ️ Turn 30/75 — turn=30 phases=['api_verified'] tests=not yet run history_len=28
📄 wrote src/agent.py
ℹ️ Turn 31/75 — turn=31 phases=['api_verified', 'source'] tests=not yet run history_len=28
📄 wrote src/init.py
ℹ️ Turn 32/75 — turn=32 phases=['api_verified', 'source'] tests=not yet run history_len=28
📄 wrote requirements.txt
ℹ️ Turn 33/75 — turn=33 phases=['api_verified', 'source'] tests=not yet run history_len=28
📄 wrote .env.example
ℹ️ Turn 34/75 — turn=34 phases=['api_verified', 'env_example', 'source'] tests=not yet run history_len=28

`run_command`: `mkdir -p tests`

exit: 0
ℹ️ Turn 35/75 — turn=35 phases=['api_verified', 'env_example', 'source'] tests=not yet run history_len=29
📄 wrote tests/init.py
ℹ️ Turn 36/75 — turn=36 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=29
📄 wrote tests/test_deepgram_connection.py
ℹ️ Turn 37/75 — turn=37 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=29
📄 wrote tests/test_livekit_connection.py
ℹ️ Turn 38/75 — turn=38 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=29
📄 wrote tests/test_integration.py
ℹ️ Turn 39/75 — turn=39 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=29
📄 wrote tests/run_tests.py
ℹ️ Turn 40/75 — turn=40 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=29

`run_command`: `pip install python-dotenv livekit-api --quiet 2>&1 | tail -3`

exit: 0


[notice] A new release of pip is available: 25.0.1 -> 26.0.1
[notice] To update, run: pip install --upgrade pip

ℹ️ Turn 41/75 — turn=41 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=30

`run_command`: `cd /workspace && python tests/test_deepgram_connection.py`

exit: 1

============================================================
Deepgram Connection Tests
============================================================

Testing plugin initialization...
  ✓ STT initialized - model: nova-3, provider: Deepgram
  ✓ TTS initialized - model: aura-2-andromeda-en, provider: Deepgram
Testing Deepgram TTS...
  ✗ TTS error: Connection error.
Testing Deepgram STT...
  ✗ STT error: Connection error.

============================================================
Tests: 1/3 passed

stderr:

failed to synthesize speech: Connection error., retrying in 0.1s
failed to synthesize speech: Connection error., retrying in 2.0s
failed to synthesize speech: Connection error., retrying in 2.0s
failed to synthesize speech: Connection error., retrying in 0.1s
failed to synthesize speech: Connection error., retrying in 2.0s
failed to synthesize speech: Connection error., retrying in 2.0s
Traceback (most recent call last):
  File "/usr/local/lib/python3.12/site-packages/livekit/plugins/deepgram/tts.py", line 203, in _run
    async with self._tts._ensure_session().post(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/livekit/plugins/deepgram/tts.py", line 156, in _ensure_session
    self._session = utils.http_context.http_session()
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/livekit/agents/utils/http_context.py", line 47, in http_session
    raise RuntimeError(
RuntimeError: Attempted to use an http session outside of a job context. This is probably because you are trying to use a plugin without using the agent worker api. You may need to create your own aiohttp.ClientSession, pass it into the plugin constructor as a kwarg, and manage its lifecycle.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/workspace/tests/test_deepgram_connection.py", line 91, in test_deepgram_stt
    async for chunk in tts.synthesize(test_phrase):
  File "/usr/local/lib/python3.12/site-packages/livekit/agents/tts/tts.py", line 355, in __anext__
    raise exc  # noqa: B904
    ^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/livekit/agents/tts/tts.py", line 199, in _traceable_main_task
    await self._main_task()
  File "/usr/local/lib/python3.12/site-packages/livekit/agents/tts/tts.py", line 292, in _main_task
    await self._run(output_emitter)
  File "/usr/local/lib/python3.12/site-packages/livekit/plugins/deepgram/tts

ℹ️ Turn 42/75 — turn=42 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=31
📄 wrote tests/test_deepgram_connection.py
ℹ️ Turn 43/75 — turn=43 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=31

`run_command`: `cd /workspace && python tests/test_deepgram_connection.py`

exit: 1

============================================================
Deepgram Connection Tests
============================================================

Testing plugin initialization...
  ✓ STT initialized - model: nova-3, provider: Deepgram
  ✓ TTS initialized - model: aura-2-andromeda-en, provider: Deepgram
Testing Deepgram TTS...
  ✓ TTS generated 89520 bytes of audio
Testing Deepgram STT...
  - Generated 18 audio frames for transcription
  ✗ STT error: 'SpeechEvent' object has no attribute 'is_final'

============================================================
Tests: 2/3 passed

stderr:

Traceback (most recent call last):
  File "/workspace/tests/test_deepgram_connection.py", line 115, in test_deepgram_stt
    if event.is_final and event.alternatives:
       ^^^^^^^^^^^^^^
AttributeError: 'SpeechEvent' object has no attribute 'is_final'

ℹ️ Turn 44/75 — turn=44 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=32

`run_command`: `python3 -c "from livekit.agents.stt import SpeechEvent; help(SpeechEvent)" 2>&1 | head -60`

exit: 0

Help on class SpeechEvent in module livekit.agents.stt.stt:

class SpeechEvent(builtins.object)
 |  SpeechEvent(type: 'SpeechEventType', request_id: 'str' = '', alternatives: 'list[SpeechData]' = <factory>, recognition_usage: 'RecognitionUsage | None' = None) -> None
 |
 |  SpeechEvent(type: 'SpeechEventType', request_id: 'str' = '', alternatives: 'list[SpeechData]' = <factory>, recognition_usage: 'RecognitionUsage | None' = None)
 |
 |  Methods defined here:
 |
 |  __eq__(self, other)
 |      Return self==value.
 |
 |  __init__(self, type: 'SpeechEventType', request_id: 'str' = '', alternatives: 'list[SpeechData]' = <factory>, recognition_usage: 'RecognitionUsage | None' = None) -> None
 |      Initialize self.  See help(type(self)) for accurate signature.
 |
 |  __repr__(self)
 |      Return repr(self).
 |
 |  ----------------------------------------------------------------------
 |  Data descriptors defined here:
 |
 |  __dict__
 |      dictionary for instance variables
 |
 |  __weakref__
 |      list of weak references to the object
 |
 |  ----------------------------------------------------------------------
 |  Data and other attributes defined here:
 |
 |  __annotations__ = {'alternatives': 'list[SpeechData]', 'recognition_us...
 |
 |  __dataclass_fields__ = {'alternatives': Field(name='alternatives',type...
 |
 |  __dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,or...
 |
 |  __hash__ = None
 |
 |  __match_args__ = ('type', 'request_id', 'alternatives', 'recognition_u...
 |
 |  recognition_usage = None
 |
 |  request_id = ''

ℹ️ Turn 45/75 — turn=45 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=33

`run_command`: `python3 -c "from livekit.agents.stt import SpeechEventType; print(list(SpeechEventType))"`

exit: 0

[<SpeechEventType.START_OF_SPEECH: 'start_of_speech'>, <SpeechEventType.INTERIM_TRANSCRIPT: 'interim_transcript'>, <SpeechEventType.PREFLIGHT_TRANSCRIPT: 'preflight_transcript'>, <SpeechEventType.FINAL_TRANSCRIPT: 'final_transcript'>, <SpeechEventType.RECOGNITION_USAGE: 'recognition_usage'>, <SpeechEventType.END_OF_SPEECH: 'end_of_speech'>]

ℹ️ Turn 46/75 — turn=46 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=34
📄 wrote tests/test_deepgram_connection.py
ℹ️ Turn 47/75 — turn=47 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=34

`run_command`: `cd /workspace && python tests/test_deepgram_connection.py`

exit: 0

============================================================
Deepgram Connection Tests
============================================================

Testing plugin initialization...
  ✓ STT initialized - model: nova-3, provider: Deepgram
  ✓ TTS initialized - model: aura-2-andromeda-en, provider: Deepgram
Testing Deepgram TTS...
  ✓ TTS generated 78960 bytes of audio
Testing Deepgram STT...
  - Generated 20 audio frames for transcription
  ✓ STT transcribed: 'The quick brown fox jumps over the lazy dog'
  ✓ Transcription accuracy check passed (8 matching words)

============================================================
All tests passed! ✓

stderr:

Deepgram keepalive task exited: Cannot write to closing transport

⚙️ Rules fired: ['R4']
ℹ️ Turn 48/75 — turn=48 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=35

`run_command`: `cd /workspace && python tests/test_livekit_connection.py`

exit: 2

============================================================
LiveKit Connection Tests
============================================================

SKIP: Missing LiveKit credentials: LIVEKIT_API_SECRET

ℹ️ Turn 49/75 — turn=49 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=36

`run_command`: `cd /workspace && python tests/test_integration.py`

exit: 0

============================================================
Integration Tests
============================================================

Testing module imports...
  ✓ All modules imported successfully
Testing Deepgram models...
  - Available model definitions loaded
  ✓ Nova-3 STT model available
  ✓ Aura-2 TTS model available
Testing AgentSession creation...
  ✓ AgentSession created successfully
    - STT model: nova-3
    - TTS model: aura-2-andromeda-en
Testing agent module...
  ✓ Agent module loaded, entrypoint found

============================================================
All tests passed! ✓

stderr:

�[0;93m2026-04-14 19:47:30.133300909 [W:onnxruntime:Default, device_discovery.cc:132 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/VMBUS:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename ""5620e0c7-8062-4dce-aeb7-520c7ef76171"" dit not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+�[m

ℹ️ Turn 50/75 — turn=50 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=37

`run_command`: `cd /workspace && python tests/run_tests.py`

exit: 0

============================================================
Deepgram Connection Tests
============================================================

Testing plugin initialization...
  ✓ STT initialized - model: nova-3, provider: Deepgram
  ✓ TTS initialized - model: aura-2-andromeda-en, provider: Deepgram
Testing Deepgram TTS...
  ✓ TTS generated 75120 bytes of audio
Testing Deepgram STT...
  - Generated 19 audio frames for transcription
  ✓ STT transcribed: 'The quick brown fox jumps over the lazy dog,'
  ✓ Transcription accuracy check passed (7 matching words)

============================================================
All tests passed! ✓
============================================================
LiveKit Connection Tests
============================================================

SKIP: Missing LiveKit credentials: LIVEKIT_API_SECRET
============================================================
Integration Tests
============================================================

Testing module imports...
  ✓ All modules imported successfully
Testing Deepgram models...
  - Available model definitions loaded
  ✓ Nova-3 STT model available
  ✓ Aura-2 TTS model available
Testing AgentSession creation...
  ✓ AgentSession created successfully
    - STT model: nova-3
    - TTS model: aura-2-andromeda-en
Testing agent module...
  ✓ Agent module loaded, entrypoint found

============================================================
All tests passed! ✓

============================================================
Running: /workspace/tests/test_deepgram_connection.py
============================================================


============================================================
Running: /workspace/tests/test_livekit_connection.py
============================================================


============================================================
Running: /workspace/tests/test_integration.py
============================================================


============================================================
TEST SUMMARY
============================================================
  test_deepgram_connection.py: PASS ✓
  test_livekit_connection.py: SKIP ⊘
  test_integration.py: PASS ✓

All tests passed!

stderr:

Deepgram keepalive task exited: Cannot write to closing transport
�[0;93m2026-04-14 19:47:45.037498162 [W:onnxruntime:Default, device_discovery.cc:132 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/VMBUS:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename ""5620e0c7-8062-4dce-aeb7-520c7ef76171"" dit not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+�[m

ℹ️ Turn 51/75 — turn=51 phases=['api_verified', 'env_example', 'source', 'tests'] tests=not yet run history_len=38
📄 wrote README.md
ℹ️ Turn 52/75 — turn=52 phases=['api_verified', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=38
📄 wrote BLOG.md
ℹ️ Turn 53/75 — turn=53 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=38

`run_command`: `pip install playwright --quiet 2>&1 | tail -2`

exit: 0

[notice] A new release of pip is available: 25.0.1 -> 26.0.1
[notice] To update, run: pip install --upgrade pip

ℹ️ Turn 54/75 — turn=54 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=39

`run_command`: `playwright install chromium 2>&1 | tail -5`

exit: 0
ℹ️ Turn 55/75 — turn=55 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=40
📄 wrote tools/create_screenshot.py
ℹ️ Turn 56/75 — turn=56 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=40

`run_command`: `cd /workspace && python tools/create_screenshot.py`

exit: 0

Screenshot saved as screenshot.png

ℹ️ Turn 57/75 — turn=57 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=41

`run_command`: `ls -la /workspace/screenshot.png`

exit: 0

-rw-r--r-- 1 root root 78394 Apr 14 19:49 /workspace/screenshot.png

ℹ️ Turn 58/75 — turn=58 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=42
ℹ️ Turn 59/75 — turn=59 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=42

`run_command`: `rm -rf /workspace/src/pycache /workspace/tools`

exit: 0
ℹ️ Turn 60/75 — turn=60 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=43
⚙️ Rules fired: ['R7']
ℹ️ Turn 61/75 — turn=61 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=43

`run_command`: `cd /workspace && python tests/run_tests.py`

exit: 0

============================================================
Deepgram Connection Tests
============================================================

Testing plugin initialization...
  ✓ STT initialized - model: nova-3, provider: Deepgram
  ✓ TTS initialized - model: aura-2-andromeda-en, provider: Deepgram
Testing Deepgram TTS...
  ✓ TTS generated 91440 bytes of audio
Testing Deepgram STT...
  - Generated 19 audio frames for transcription
  ✓ STT transcribed: 'The quick brown fox jumps over the lazy dog.'
  ✓ Transcription accuracy check passed (8 matching words)

============================================================
All tests passed! ✓
============================================================
LiveKit Connection Tests
============================================================

SKIP: Missing LiveKit credentials: LIVEKIT_API_SECRET
============================================================
Integration Tests
============================================================

Testing module imports...
  ✓ All modules imported successfully
Testing Deepgram models...
  - Available model definitions loaded
  ✓ Nova-3 STT model available
  ✓ Aura-2 TTS model available
Testing AgentSession creation...
  ✓ AgentSession created successfully
    - STT model: nova-3
    - TTS model: aura-2-andromeda-en
Testing agent module...
  ✓ Agent module loaded, entrypoint found

============================================================
All tests passed! ✓

============================================================
Running: /workspace/tests/test_deepgram_connection.py
============================================================


============================================================
Running: /workspace/tests/test_livekit_connection.py
============================================================


============================================================
Running: /workspace/tests/test_integration.py
============================================================


============================================================
TEST SUMMARY
============================================================
  test_deepgram_connection.py: PASS ✓
  test_livekit_connection.py: SKIP ⊘
  test_integration.py: PASS ✓

All tests passed!

stderr:

Deepgram keepalive task exited: Cannot write to closing transport
�[0;93m2026-04-14 19:49:51.822979668 [W:onnxruntime:Default, device_discovery.cc:132 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/VMBUS:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename ""5620e0c7-8062-4dce-aeb7-520c7ef76171"" dit not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+�[m

ℹ️ Turn 62/75 — turn=62 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=44

`run_command`: `cd /workspace && ls -la && echo "---" && cat .env.example`

exit: 0

total 128
drwxr-xr-x 4 1001 1001  4096 Apr 14 19:49 .
drwxr-xr-x 1 root root  4096 Apr 14 19:49 ..
-rw------- 1 1001 1001   239 Apr 14 19:45 .env.example
-rw------- 1 1001 1001 12651 Apr 14 19:48 BLOG.md
-rw------- 1 1001 1001  4649 Apr 14 19:48 README.md
-rw------- 1 1001 1001   296 Apr 14 19:45 requirements.txt
-rw-r--r-- 1 root root 78394 Apr 14 19:49 screenshot.png
drwxr-xr-x 3 root root  4096 Apr 14 19:49 src
drwxr-xr-x 2 root root  4096 Apr 14 19:47 tests
---
# Deepgram API key for speech-to-text and text-to-speech
DEEPGRAM_API_KEY=

# LiveKit server connection
LIVEKIT_URL=wss://your-livekit-server.livekit.cloud
LIVEKIT_API_KEY=
LIVEKIT_API_SECRET=

# OpenAI API key for the LLM
OPENAI_API_KEY=

ℹ️ Turn 63/75 — turn=63 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=45

`run_command`: `grep -r "DEEPGRAM_API_KEY\|LIVEKIT\|OPENAI" /workspace/src/ --include="*.py" | head -5`

exit: 0
ℹ️ Turn 64/75 — turn=64 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=46
ℹ️ Turn 65/75 — turn=65 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=46
ℹ️ Turn 66/75 — turn=66 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=46

`run_command`: `rm -rf /workspace/src/pycache`

exit: 0
ℹ️ Turn 67/75 — turn=67 phases=['api_verified', 'blog', 'env_example', 'readme', 'source', 'tests'] tests=not yet run history_len=47
ℹ️ Agent signalled completion after 67 turns — constraints verified ✓

Relates to #225

lukeocodes · 2026-04-14T20:16:50Z

Closing to rebuild with updated pipeline (clean PR body, needs-credentials label).

feat(examples): add 540-livekit-voice-agent-python

c9a85cd

Relates to #225

github-actions bot added type:example New example app automated Opened by the build pipeline labels Apr 14, 2026

github-actions bot mentioned this pull request Apr 14, 2026

livekit guides to demos #225

Open

lukeocodes closed this Apr 14, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(examples): add 540-livekit-voice-agent-python#231

feat(examples): add 540-livekit-voice-agent-python#231
github-actions[bot] wants to merge 1 commit intomainfrom
example/540-livekit-voice-agent-python

github-actions bot commented Apr 14, 2026

Uh oh!

lukeocodes commented Apr 14, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

github-actions bot commented Apr 14, 2026

Build log

Build log: example 540-livekit-voice-agent-python

run_command: echo "DEEPGRAM_API_KEY is set: ${DEEPGRAM_API_KEY:+yes}" && echo "LIVEKIT_API_KEY is set: ${LIVEKIT_API_KEY:+yes}" && echo "LIVEKIT_API_SECRET is set: ${LIVEKIT_API_SECRET:+yes}" && echo "LIVEKIT_URL is set: ${LIVEKIT_URL:+yes}"

run_command: deepgram --help 2>/dev/null || echo "deepgram CLI not available"

run_command: context7 search "livekit agents python deepgram voice"

run_command: pip install livekit-agents --dry-run 2>&1 | head -20

run_command: pip search livekit-plugins-deepgram 2>&1 || pip index versions livekit-plugins-deepgram 2>&1 || pip install livekit-plugins-deepgram==9999 2>&1 | head -20

run_command: cat /workspace/../030-livekit-agents-python/README.md 2>/dev/null || echo "Looking for existing example..."

run_command: ls -la /workspace/../ 2>/dev/null | head -20 || echo "can't list parent"

run_command: pip install livekit-agents livekit-plugins-deepgram livekit-plugins-silero --quiet 2>&1 | tail -5

run_command: python3 -c "import livekit.agents; print(livekit.agents.__version__)"

run_command: python3 -c "from livekit.plugins import deepgram; print(dir(deepgram))"

run_command: python3 -c "from livekit.plugins.deepgram import STT, TTS; help(STT)" 2>&1 | head -80

run_command: python3 -c "from livekit.plugins.deepgram import TTS; help(TTS)" 2>&1 | head -80

run_command: python3 -c "from livekit.agents import AgentSession, Agent; help(AgentSession)" 2>&1 | head -100

run_command: python3 -c "from livekit.agents import Agent; help(Agent)" 2>&1 | head -60

run_command: python3 -c "from livekit.agents import cli; help(cli)" 2>&1 | head -50

run_command: python3 -c "from livekit.agents import cli; print(dir(cli))"

run_command: python3 -c "from livekit.agents.cli import run_app; help(run_app)" 2>&1 | head -100

run_command: python3 -c "from livekit.agents import AgentServer; help(AgentServer)" 2>&1 | head -50

run_command: pip install livekit-plugins-openai --quiet 2>&1 | tail -3

run_command: python3 -c "from livekit.plugins import openai; print(dir(openai))"

run_command: pip index versions livekit-plugins-silero 2>&1 | head -5

run_command: echo "OPENAI_API_KEY is set: ${OPENAI_API_KEY:+yes}"

run_command: pip search livekit-plugins 2>&1 || pip install livekit-plugins-anthropic==9999 2>&1 | head -15

run_command: echo "ANTHROPIC_API_KEY is set: ${ANTHROPIC_API_KEY:+yes}"

run_command: pip install livekit-plugins-google==9999 2>&1 | head -5

run_command: echo "GOOGLE_API_KEY is set: ${GOOGLE_API_KEY:+yes}"

run_command: echo "GROQ_API_KEY is set: ${GROQ_API_KEY:+yes}"

run_command: env | grep -i "_API_KEY\|_API_SECRET\|_URL" | sed 's/=.*/=***/' | sort

run_command: mkdir -p tests

run_command: pip install python-dotenv livekit-api --quiet 2>&1 | tail -3

run_command: cd /workspace && python tests/test_deepgram_connection.py

run_command: cd /workspace && python tests/test_deepgram_connection.py

run_command: python3 -c "from livekit.agents.stt import SpeechEvent; help(SpeechEvent)" 2>&1 | head -60

run_command: python3 -c "from livekit.agents.stt import SpeechEventType; print(list(SpeechEventType))"

run_command: cd /workspace && python tests/test_deepgram_connection.py

run_command: cd /workspace && python tests/test_livekit_connection.py

run_command: cd /workspace && python tests/test_integration.py

run_command: cd /workspace && python tests/run_tests.py

run_command: pip install playwright --quiet 2>&1 | tail -2

run_command: playwright install chromium 2>&1 | tail -5

run_command: cd /workspace && python tools/create_screenshot.py

run_command: ls -la /workspace/screenshot.png

run_command: rm -rf /workspace/src/__pycache__ /workspace/tools

run_command: cd /workspace && python tests/run_tests.py

run_command: cd /workspace && ls -la && echo "---" && cat .env.example

run_command: grep -r "DEEPGRAM_API_KEY\|LIVEKIT\|OPENAI" /workspace/src/ --include="*.py" | head -5

run_command: rm -rf /workspace/src/__pycache__

Uh oh!

lukeocodes commented Apr 14, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

`run_command`: `echo "DEEPGRAM_API_KEY is set: ${DEEPGRAM_API_KEY:+yes}" && echo "LIVEKIT_API_KEY is set: ${LIVEKIT_API_KEY:+yes}" && echo "LIVEKIT_API_SECRET is set: ${LIVEKIT_API_SECRET:+yes}" && echo "LIVEKIT_URL is set: ${LIVEKIT_URL:+yes}"`

`run_command`: `deepgram --help 2>/dev/null || echo "deepgram CLI not available"`

`run_command`: `context7 search "livekit agents python deepgram voice"`

`run_command`: `pip install livekit-agents --dry-run 2>&1 | head -20`

`run_command`: `pip search livekit-plugins-deepgram 2>&1 || pip index versions livekit-plugins-deepgram 2>&1 || pip install livekit-plugins-deepgram==9999 2>&1 | head -20`

`run_command`: `cat /workspace/../030-livekit-agents-python/README.md 2>/dev/null || echo "Looking for existing example..."`

`run_command`: `ls -la /workspace/../ 2>/dev/null | head -20 || echo "can't list parent"`

`run_command`: `pip install livekit-agents livekit-plugins-deepgram livekit-plugins-silero --quiet 2>&1 | tail -5`

`run_command`: `python3 -c "import livekit.agents; print(livekit.agents.version)"`

`run_command`: `python3 -c "from livekit.plugins import deepgram; print(dir(deepgram))"`

`run_command`: `python3 -c "from livekit.plugins.deepgram import STT, TTS; help(STT)" 2>&1 | head -80`

`run_command`: `python3 -c "from livekit.plugins.deepgram import TTS; help(TTS)" 2>&1 | head -80`

`run_command`: `python3 -c "from livekit.agents import AgentSession, Agent; help(AgentSession)" 2>&1 | head -100`

`run_command`: `python3 -c "from livekit.agents import Agent; help(Agent)" 2>&1 | head -60`

`run_command`: `python3 -c "from livekit.agents import cli; help(cli)" 2>&1 | head -50`

`run_command`: `python3 -c "from livekit.agents import cli; print(dir(cli))"`

`run_command`: `python3 -c "from livekit.agents.cli import run_app; help(run_app)" 2>&1 | head -100`

`run_command`: `python3 -c "from livekit.agents import AgentServer; help(AgentServer)" 2>&1 | head -50`

`run_command`: `pip install livekit-plugins-openai --quiet 2>&1 | tail -3`

`run_command`: `python3 -c "from livekit.plugins import openai; print(dir(openai))"`

`run_command`: `pip index versions livekit-plugins-silero 2>&1 | head -5`

`run_command`: `echo "OPENAI_API_KEY is set: ${OPENAI_API_KEY:+yes}"`

`run_command`: `pip search livekit-plugins 2>&1 || pip install livekit-plugins-anthropic==9999 2>&1 | head -15`

`run_command`: `echo "ANTHROPIC_API_KEY is set: ${ANTHROPIC_API_KEY:+yes}"`

`run_command`: `pip install livekit-plugins-google==9999 2>&1 | head -5`

`run_command`: `echo "GOOGLE_API_KEY is set: ${GOOGLE_API_KEY:+yes}"`

`run_command`: `echo "GROQ_API_KEY is set: ${GROQ_API_KEY:+yes}"`

`run_command`: `env | grep -i "_API_KEY\|_API_SECRET\|_URL" | sed 's/=.*/=***/' | sort`

`run_command`: `mkdir -p tests`

`run_command`: `pip install python-dotenv livekit-api --quiet 2>&1 | tail -3`

`run_command`: `cd /workspace && python tests/test_deepgram_connection.py`

`run_command`: `cd /workspace && python tests/test_deepgram_connection.py`

`run_command`: `python3 -c "from livekit.agents.stt import SpeechEvent; help(SpeechEvent)" 2>&1 | head -60`

`run_command`: `python3 -c "from livekit.agents.stt import SpeechEventType; print(list(SpeechEventType))"`

`run_command`: `cd /workspace && python tests/test_deepgram_connection.py`

`run_command`: `cd /workspace && python tests/test_livekit_connection.py`

`run_command`: `cd /workspace && python tests/test_integration.py`

`run_command`: `cd /workspace && python tests/run_tests.py`

`run_command`: `pip install playwright --quiet 2>&1 | tail -2`

`run_command`: `playwright install chromium 2>&1 | tail -5`

`run_command`: `cd /workspace && python tools/create_screenshot.py`

`run_command`: `ls -la /workspace/screenshot.png`

`run_command`: `rm -rf /workspace/src/pycache /workspace/tools`

`run_command`: `cd /workspace && python tests/run_tests.py`

`run_command`: `cd /workspace && ls -la && echo "---" && cat .env.example`

`run_command`: `grep -r "DEEPGRAM_API_KEY\|LIVEKIT\|OPENAI" /workspace/src/ --include="*.py" | head -5`

`run_command`: `rm -rf /workspace/src/pycache`