Update LiveKit dependencies (non-major) to v1.6.7

[renovate]

ℹ️ Note

This PR body was truncated due to platform limits.

This PR contains the following updates:

Package	Change	Age	Confidence
livekit-agents	==1.2.7 → ==1.6.7
livekit-plugins-deepgram	==1.2.7 → ==1.6.7
livekit-plugins-openai	==1.2.7 → ==1.6.7
livekit-plugins-silero	==1.2.7 → ==1.6.7

---

Release Notes

livekit/agents (livekit-agents)

v1.6.7: @​livekit/agents@1.6.7

Compare Source

What's Changed

• fix(core): reset user away timer on final STT transcript AGT-3149 by @​chenghao-mou in #​6478
• Fix stripping of Gemma reasoning markers in streamed output by @​C0deRatoR in #​6398
• feat: Spatius avatar plugin by @​3DRX in #​5821
• fix(avatar): skip removal if avatar not found in the room by @​longcw in #​6486
• fix(endpointing): keep max_delay constant in dynamic endpointing by @​chenghao-mou in #​6265
• fix(stt): don't clear the primary speaker on an unattributed segment by @​chuenchen309 in #​6460
• fix(drive-thru): assume quantity of one when unstated to avoid count confirmation by @​u9g in #​6463
• fix: make AMD settle on endpointing backstop by default by @​chenghao-mou in #​6496
• fix(voice): include function name in FunctionCallOutput protos by @​u9g in #​6491
• feat: tag simulation and redaction telemetry by @​chenghao-mou in #​6497
• Update READMEs to use gemma by @​keepingitneil in #​6502
• fix(voice): don't drop the realtime turn when the chat-ctx push times out by @​ByteMaster-1 in #​6445
• feat(voice): expose realtime provider request IDs by @​yaodong-shen in #​6501
• Add 'zh-HK' to DeepgramLanguages by @​devb-enp in #​6507
• (gnani): Update authors in pyproject.toml by @​tinalenguyen in #​6489
• Allow local VAD inference to use a dedicated executor by @​basnijholt in #​6498
• feat(fishaudio): add normalize_loudness and generation-tuning TTS options by @​adrian-cowham in #​6520
• fix(voice): don't hold session.run() open on update_agent handoffs with long-lived on_enter by @​k-zaher in #​6500
• plugins(krisp): add voice isolation telephony mode by @​lukasIO in #​6510
• Support adaptive interruption for realtime models by @​longcw in #​6488
• feat(inference): add inference.AvatarSession (avatar provisioning via the gateway) by @​shawnfeldman in #​6492
• (inference): add fish audio tts by @​tinalenguyen in #​6529
• slng: rewrite plugin around the Unmute Bridge (v2) by @​metehan-slng in #​6442
• feat(speechify): streaming TTS with word-level timestamps via official SDK by @​luke-speechify in #​6327
• fix(openai): recover from stale pooled Responses WebSocket connections by @​longcw in #​6523
• prewarm LLM to skip connection establishment by @​davidzhao in #​6484
• fix(otel): skip otel upload error logs if it's disabled by the project setting AGT-3015 by @​chenghao-mou in #​6167
• chore(spatius): bump spatius to 1.0.4 by @​3DRX in #​6536
• fix(livekit-agents): require livekit-protocol>=1.1.21 for MetricsRecordingHeader.simulated by @​u9g in #​6541
• (xai tts): add new voices by @​tinalenguyen in #​6543
• fix(speechify): restore caller header, close owned httpx client by @​luke-speechify in #​6542
• fix(plugins-aws): Polly TTS silent on aiobotocore >=3.8.0 (AioStreamingBody has no .content) by @​Gooh456 in #​6540
• feat(telemetry): include agent_name as an OTLP resource attribute by @​jmcclanahan in #​6539
• livekit-agents@​1.6.7 by @​github-actions[bot] in #​6532

New Contributors

• @​C0deRatoR made their first contribution in #​6398
• @​3DRX made their first contribution in #​5821
• @​chuenchen309 made their first contribution in #​6460
• @​yaodong-shen made their first contribution in #​6501
• @​basnijholt made their first contribution in #​6498
• @​k-zaher made their first contribution in #​6500
• @​shawnfeldman made their first contribution in #​6492
• @​luke-speechify made their first contribution in #​6327
• @​Gooh456 made their first contribution in #​6540
• @​jmcclanahan made their first contribution in #​6539

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.6.6...livekit-agents@1.6.7

v1.6.6

Compare Source

What's Changed

• livekit-plugins-fishaudio: pooled websocket reuse + prebuffer to fix cold-start crackle by @​cshape in #​6368
• rename reload-addr flag to cli-addr by @​davidzhao in #​6376
• fix(drive-thru): instruct agent to add stated quantities without confirming by @​u9g in #​6383
• elevenlabs: deprecate STT model_id in favor of model by @​u9g in #​6378
• feat(phonic): add additional params support by @​qionghuang6 in #​6384
• fix(openai): omit unset fields when serializing Responses WS requests by @​markine in #​6385
• fix(voice): preserve completed tool results in chat context on interruption by @​hello-rory in #​6349
• elevenlabs: log trace id for failed requests by @​longcw in #​6396
• frontdesk: end-to-end simulation example with dynamic tool mocking by @​theomonnom in #​6089
• (xai tts): adjust websocket deadline by @​tinalenguyen in #​6393
• (examples revamp): remove redundant files by @​tinalenguyen in #​5523
• Add LiveKit cloud auth to krisp viva plugin by @​lukasIO in #​5914
• feat(soniox): add endpoint latency adjustment parameter by @​FLoppix in #​6400
• fix(assemblyai): stop forcing continuous_partials=true for U3-Pro models by @​u9g in #​6401
• (inference xai): add speed param by @​tinalenguyen in #​6394
• fix: Instructions.resolve_template double-renders under render(modality) by @​longcw in #​6414
• fix(openai): stop realtime reconnect loop on fatal server errors (insufficient_quota, invalid_api_key) by @​ByteMaster-1 in #​6352
• fix(openai): include model in WebSocket URLs by @​toubatbrian in #​6403
• livekit-plugins-fishaudio: drop the startup prebuffer (fish now streams the opening chunks smoothly) by @​cshape in #​6423
• (google llm): support mixed tool use by @​tinalenguyen in #​6416
• fix(google): gate mixed tool use to the Gemini 3 Developer API + build tool config in chat() by @​tinalenguyen in #​6426
• (google llm): strip thought summaries by @​tinalenguyen in #​6428
• feat(fishaudio): expose temperature, top_p, bitrate, and normalize TTS options by @​cshape in #​6456
• feat(assemblyai): add language_codes steering param, enable agent_context_carryover by default by @​dlange-aai in #​6443
• chore(deps): update dependency mcp to v1.28.1 [security] by @​renovate[bot] in #​6457
• chore(deps): update dependency nltk to v3.10.0 [security] by @​renovate[bot] in #​6458
• fix(deps): update dependency json-repair to v0.60.1 [security] by @​renovate[bot] in #​6413
• fix: avoid mid-word splits in text_transforms.replace by @​longcw in #​6439
• fix(google): close realtime output streams on generation_complete by @​longcw in #​6431
• feat(mcp): per-tool options and progress update for MCPToolset by @​longcw in #​5694
• feat(agent): add update_models() to swap STT/VAD/LLM/TTS at runtime by @​longcw in #​6440
• fix(voice): recover STT stream after unrecoverable errors by @​longcw in #​6418
• chore(pii): add job_id in MetricsRecordingHeader by @​chenghao-mou in #​6326
• voice: add RemoteSession.get_framework_info() client wrapper by @​u9g in #​6464
• (deepgram stt): fix reconnect on send-side socket drops instead of crashing the session by @​tinalenguyen in #​6429
• (expressive mode): add to update_options by @​tinalenguyen in #​6367
• fix(plugins-aws): depend on aiobotocore directly instead of aioboto3 by @​xavieralmendros-aily in #​6438
• (xai tts inference): add optimize_streaming_latency param, fix typecheck by @​tinalenguyen in #​6469
• fix(stt): raise retryable APIConnectionError on mid-send socket drops by @​tinalenguyen in #​6467
• feat(smallestai): use Waves continuous WebSocket streaming protocol for TTS by @​harshitajain165 in #​6363
• fix(soniox): sentence-buffer tts input, rotate streams on idle and age by @​chenghao-mou in #​6434
• (assemblyai inference): support agent_context_carryover by @​tinalenguyen in #​6466
• feat(gnani): update TTS voices/models and STT sample rates + streaming toggle by @​tinalenguyen in #​6468
• assemblyai: make agent_context_carryover opt-in (off by default) by @​tinalenguyen in #​6471
• feat(stt): add stt_context_options for keyterms and chat context by @​longcw in #​6475
• livekit-agents@​1.6.6 by @​github-actions[bot] in #​6476

New Contributors

• @​markine made their first contribution in #​6385
• @​hello-rory made their first contribution in #​6349
• @​xavieralmendros-aily made their first contribution in #​6438

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.6.5...livekit-agents@1.6.6

v1.6.5: @​livekit/agents@1.6.5

Compare Source

What's Changed

• fix(google): broaden _is_gemini_3_flash_model to cover all Gemini 3 Flash variants by @​balazssandor in #​6210
• fix(assemblyai): respect mode preset for turn-silence; remap deprecated u3-pro by @​dlange-aai in #​6220
• Move frame processor url/token/stream info to client sdk by @​1egoman in #​5867
• fix(worker): exit when connection fails after max retries by @​longcw in #​6103
• fix(voice): don't re-issue in-flight tool calls on a new turn by @​longcw in #​6227
• (phonic): Use stream_ahead_of_real_time mode for Phonic WebSocket and sample rate 24000 by @​qionghuang6 in #​6236
• fix(anthropic): use non-whitespace trailing dummy to satisfy provider validation by @​JasonGordonD in #​6216
• fix(deepgram): sync missing STT/TTS API params, fix duplicate languages, add TTS model list by @​Namit1867 in #​5362
• chore(deps): update dependency msgpack to v1.2.1 [security] by @​renovate[bot] in #​6243
• docs: fix hotel_receptionist architecture section after refactor by @​detail-app[bot] in #​6230
• fix(deepgram): remove duplicate redact param by @​longcw in #​6245
• fix(openai): delete trimmed empty audio items by @​he-yufeng in #​5887
• fix(openai realtime): discard orphaned response after interrupt or timeout by @​longcw in #​6244
• fix CI due to realtime_model test by @​davidzhao in #​6261
• fix: preserve transcription item id by @​nightcityblade in #​6120
• fix(openai): skip realtime truncate when no audio was played fixes : #​6157 by @​ByteMaster-1 in #​6158
• feat(assemblyai): add language_code streaming param for language steering by @​gsharp-aai in #​6219
• fix(xai): emit interim transcripts for in-progress status in realtime by @​ByteMaster-1 in #​6272
• conversation-aware STT recognition (keyterms + chat context) by @​longcw in #​6039
• Allow LemonSlice avatars to join third party meeting platforms by @​aweitz in #​6233
• (xai realtime): pass voice parameter by @​tinalenguyen in #​6276
• feat(certifi): user certifi as a fallback when system store is not present by @​chenghao-mou in #​6231
• chore(eot): update log message about slow stt transcript by @​chenghao-mou in #​6179
• fix(fallback_adapter): update stream start time offset in stt fallbackadapter by @​chenghao-mou in #​6246
• feat(voice): stream tool call status events by @​longcw in #​6100
• set default shutdown reason by @​davidzhao in #​6284
• bump worker drain timeout to 1hr by @​chenghao-mou in #​6286
• feat(rime): support speed_alpha for arcana and coda models by @​balazssandor in #​6287
• feat(deepgram): apply Flux option updates in-band by @​longcw in #​6277
• feat: realtime model fallback adapter by @​longcw in #​6282
• update keyterm detection default model by @​longcw in #​6289
• fix(eot): tighten eot cancellation by speech acitivity by @​chenghao-mou in #​6274
• silero: use onnx when explicitely using silero.VAD.load() by @​theomonnom in #​6290
• fix(openai): emit start/end of speech events for realtime STT AGT-3051 by @​chenghao-mou in #​6297
• feat(smallestai): word_timestamps for TTS, v4 STT endpoints, eou_timeout fix by @​harshitajain165 in #​5953
• fix(aws): nest endpointingSensitivity under turnDetectionConfiguration for Nova Sonic 2 by @​St-Luciferr in #​6201
• switch to blingfire tokenizer by default by @​davidzhao in #​6301
• feat(plugins-tavus): rename replica/persona to face/pal by @​carolin-tavus in #​6239
• fix(openai): avoid mutating raw tool schemas for Responses by @​VectorPeak in #​6305
• feat(openai): add temperature parameter to STT by @​longcw in #​6307
• fix(liveavatar): preserve provider status errors by @​VectorPeak in #​6306
• fix incorrect attribution of latency in tts.ttfb by @​davidzhao in #​6300
• update livekit to 1.1.13 by @​davidzhao in #​6311
• fix(voice): race between session.run and an in-flight AgentTask handoff by @​longcw in #​6315
• fix(voice): resolve video task leak in AgentSession by @​Mayank-MSJ-Singh in #​6318
• Add remove() method to ChatContext for flexible item removal by @​3eid in #​5131
• fix: expose speech generation errors via SpeechHandle.exception() by @​longcw in #​6304
• chore(deps): update actions/checkout action to v7 by @​renovate[bot] in #​6325
• fix(test): stop false-positive "Audio data isn't PCM" failures in TTS tests by @​u9g in #​6331
• feat(anam) add director notes config by @​sr-anam in #​6267
• fix(aws): preserve image format for Bedrock images by @​VectorPeak in #​6328
• (feat) 11Labs STT : Add enable_logging by @​Ludobaka in #​6273
• feat: expressive mode by @​tinalenguyen in #​6116
• Add connection pooling for xAI TTS by @​godququ5-code in #​6317
• voice: output retries for run(output_type=...) by @​theomonnom in #​6080
• disable openai sdk max_retries by @​chenghao-mou in #​6341
• Allow direct image uploads for LemonSlice avatars by @​jp-lemon in #​6343
• remove structured LLM output from voice pipeline by @​tinalenguyen in #​6345
• remove livekit-blingfire (moved) by @​theomonnom in #​6320
• (expressive mode): expression marker dialect by @​tinalenguyen in #​6347
• (expressive mode): fix markup tags impacting pacing by @​tinalenguyen in #​6350
• feat(core): add plain_text_content property for ChatMessage by @​chenghao-mou in #​6348
• fix(tts): restart ChunkedStream retries under a fresh request_id by @​u9g in #​6346
• send ServerInfo to CLI host on start by @​davidzhao in #​6260
• docs: fix livekit-durable README title copied from removed blingfire plugin by @​detail-app[bot] in #​6355
• Feat/updating documentation for Gnani by @​Gnani-AI-Mintlify in #​6004
• feat(soniox): support speed parameter in TTS by @​yui0303 in #​6339
• fix(azure): recover from STT error cancellations instead of hanging by @​apoorva-01 in #​6362
• fix: honor IGNORE_ON_ENTER for tools inside a Toolset by @​longcw in #​6361
• fix: send DTMF to the active session room by @​longcw in #​6360
• docs(soniox): add speed parameter to TTS example in README by @​detail-app[bot] in #​6366
• (expressive mode): hide and remove from examples by @​tinalenguyen in #​6370
• fix(core): flip text content fields AGT-3103 by @​chenghao-mou in #​6353
• livekit-agents@​1.6.5 by @​github-actions[bot] in #​6371

New Contributors

• @​balazssandor made their first contribution in #​6210
• @​1egoman made their first contribution in #​5867
• @​JasonGordonD made their first contribution in #​6216
• @​ByteMaster-1 made their first contribution in #​6158
• @​aweitz made their first contribution in #​6233
• @​St-Luciferr made their first contribution in #​6201
• @​carolin-tavus made their first contribution in #​6239
• @​VectorPeak made their first contribution in #​6305
• @​Mayank-MSJ-Singh made their first contribution in #​6318
• @​3eid made their first contribution in #​5131
• @​godququ5-code made their first contribution in #​6317
• @​yui0303 made their first contribution in #​6339
• @​apoorva-01 made their first contribution in #​6362

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.6.4...livekit-agents@1.6.5

v1.6.4

Compare Source

What's Changed

Bug fixes

Turn Detector

• disable retry for eot errors by @​chenghao-mou in #​6195
• disable eot connection error retry too by @​chenghao-mou in #​6196

Turn handling/committing

• fix(agents): keep STT input anchor on the pipeline across handoff by @​chenghao-mou in #​6207

[!WARNING]
users are advised to upgrade if you use agent handoff with STT and are on 1.5.14 ~ 1.6.3

Other updates

Core

• chore(deps): update livekit dependency by @​chenghao-mou in #​6194
• chore(example): update otel trace example by @​chenghao-mou in #​6178
• chore(dep): update local-inference dep by @​chenghao-mou in #​6214
• update livekit version by @​chenghao-mou in #​6217
• Hotel Receptionist Scenario Expansion by @​ShayneP in #​6186

Plugins

• chore(elevenlabs): update language_code doc to remove error case by @​inickt in #​6197
• (xai realtime): set text modality by @​tinalenguyen in #​6198
• fix(google): clean up STT input frame task by @​gyx09212214-prog in #​6193
• feat: add Protoface avatar plugin by @​protoface-team in #​6164

Version update

• livekit-agents@​1.6.4 by @​github-actions[bot] in #​6218

New Contributors

• @​gyx09212214-prog made their first contribution in #​6193
• @​protoface-team made their first contribution in #​6164

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.6.3...livekit-agents@1.6.4

v1.6.3

Compare Source

What's Changed

• fix(slng): expose speed in update_options by @​adityajha2005 in #​6175
• fix(google): reject tool calls when tool_choice="none" in realtime by @​longcw in #​6166
• fix(eot): restore timeout behavior for eot inference by @​chenghao-mou in #​6188
• (assembly ai): add inference params by @​tinalenguyen in #​6185
• livekit-agents@​1.6.3 by @​github-actions[bot] in #​6189

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.6.2...livekit-agents@1.6.3

v1.6.2

Compare Source

What's Changed

• feat(assemblyai): add universal-3-5-pro (now default) and Voice Focus streaming params by @​dlange-aai in #​6119
• feat: add Gemini 3.1 flash TTS support, implement streaming response … by @​thorwebdev in #​6134
• (google): fix formatting by @​tinalenguyen in #​6142
• chore(examples): update livekit-agents version to >=1.6 in requirements.txt by @​Bobronium in #​6150
• fix(openai): default reasoning_effort to 'none' for mini/nano models by @​sarvesh4396 in #​6148
• feat(assemblyai): add streaming mode (latency/accuracy preset) param by @​dlange-aai in #​6156
• feat(fishaudio): add speed/volume prosody params and default to s2.1-pro by @​cshape in #​6159
• docs: add missing google_gemini_tts.py example to voice agents README by @​detail-app[bot] in #​6155
• feat(soniox): support stt-rt-v5 with endpoint_sensitivity option by @​mihafabcic-soniox in #​6126
• fix(xai): expose TTS websocket options by @​nightcityblade in #​6095
• (phonic) hotfix: add a config that lists tools with forbid-speech-after-tool-call by @​arunwpm-work in #​6143
• fix(workflows): normalize two-digit years in GetDOBTask by @​he-yufeng in #​6124
• fix(azure): emit STT recognition usage metrics by @​nightcityblade in #​6161
• (phonic): reuse ws connection across handoffs by @​tinalenguyen in #​6163
• (google): surface context exhaustion error by @​tinalenguyen in #​6144
• (inference): add assembly ai model by @​tinalenguyen in #​6162
• docs(phonic): add forbid_speech_after_tool_call to README config table by @​detail-app[bot] in #​6168
• feat(phonic): handle user text input by @​tinalenguyen in #​6169
• livekit-agents@​1.6.2 by @​github-actions[bot] in #​6170

New Contributors

• @​thorwebdev made their first contribution in #​6134
• @​sarvesh4396 made their first contribution in #​6148

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.6.1...livekit-agents@1.6.2

v1.6.1

Compare Source

Introducing LiveKit Turn Detector v1.0

Determining when the agent should respond is a delicate balance: responding too early can interrupt the user, while responding too late introduces unnecessary latency and awkward silence. We solve this with our state-of-the-art turn detector, which leverages both audio and text semantics to identify the optimal moment to respond.

Read more about our turn detector in our blog.

What's Changed

• fix(ipc): close supervised process when start() is cancelled by @​longcw in #​6051
• simulation: participant-sync race, fake-job metadata fallback, no audio recording by @​theomonnom in #​6069
• fix(voice): deadlock when session closes while awaiting an AgentTask by @​longcw in #​6071
• fix(voice): add timeout fallback to transcript segment aclose by @​longcw in #​6029
• add codeowners by @​davidzhao in #​6072
• (phonic) Add headers to conversations made through livekit-agents-py by @​arunwpm-work in #​6023
• chore(ipc): increase memory warning threshold by @​Bobronium in #​6073
• chore(deps): update av to 17.1.0 by @​kyle-livekit in #​6091
• fix(clova): stop _recognize_impl from mutating instance language as a side effect by @​adityajha2005 in #​6097
• Drop unused agent_id kwarg from AgentConfigUpdate by @​longcw in #​6090
• fix(avatar): preserve audio wrappers across avatar hot-swaps by @​longcw in #​5863
• fix(deps): update livekit deps to v1.1.9 by @​renovate[bot] in #​6102
• inference(stt): remove cartesia/ink-2-latest from Cartesia STT type hints by @​adrian-cowham in #​6117
• fix: honor per-agent preemptive_generation options by @​longcw in #​6114
• fix(azure): flush buffered audio on end_input to avoid truncated STT by @​jose-speak in #​6118
• fix(worker): fix timeout option overrides and update initialize_process_timeout by @​parkky21 in #​6121
• fix(voice): include conversation items added during tool execution in reply context by @​ShayneP in #​6084
• feat(llm): list available tools in unknown-function error output by @​ShayneP in #​6085
• fix(workflows): normalize placeholder strings in GetNameTask arguments by @​ShayneP in #​6086
• fix(voice): skip end-of-turn metrics on stale/out-of-order speaking anchor by @​anshulkulhari7 in #​6098
• feat(eot): add audio models AGT-2520 by @​chenghao-mou in #​4722
• Improve hotel receptionist example: policies, card collection, richer DB by @​ShayneP in #​6087
• feat(core): allow non-primary sessions without job_ctx by @​Bobronium in #​6137
• Revert "feat(core): allow non-primary sessions without job_ctx" by @​Bobronium in #​6139
• livekit-agents@​1.6.1 by @​github-actions[bot] in #​6140

New Contributors

• @​kyle-livekit made their first contribution in #​6091
• @​parkky21 made their first contribution in #​6121
• @​anshulkulhari7 made their first contribution in #​6098

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.6.0...livekit-agents@1.6.1

v1.6.0

Introducing Asynchronous Tools

When a long-running tool is in progress, the user will be met with silence until it completes. Asynchronous tools can hand control back to the LLM before it finishes, streaming updates into the conversation as it progresses.

class TravelAgent(Agent):
    @​llm.function_tool(flags=llm.ToolFlag.CANCELLABLE, on_duplicate="confirm")
    async def book_flight(self, ctx: RunContext, origin: str, destination: str, date: str) -> str:
        """Book a flight."""
        # First update is delivered immediately and releases control back to the LLM,
        # so the agent can say e.g. "Sure, searching flights to Tokyo — this'll take a minute."
        await ctx.update(
            f"Searching flights from {origin} to {destination} on {date}. "
            "This will take a couple of minutes."
        )

        await asyncio.sleep(30)  # simulate long-running work

        # Later updates are coalesced into a deferred reply, delivered when the agent
        # is idle: "Good news — best price is $289 on Delta, confirming now."
        await ctx.update("Found 3 options. Best price $289 on Delta. Confirming now.")

        await asyncio.sleep(40)  # simulate more long-running work

        # The final return is also delivered when the agent is idle.
        return f"Booked! Confirmation FL-{random.randint(100000, 999999)}."

Filler phrases

Use filler words to break long silences via ctx.with_filler(). You can set delay to the amount of quiet seconds per interval for when the filler words are played. You can also rotate between phrases, set interval to configure the cooldown between filler fires.

        followups = [
            "Almost there, just confirming.",
            "Still working on it, won't be long.",
            "Hang tight — almost done.",
        ]
        async with ctx.with_filler(
            lambda step: followups[step], delay=5, interval=10, max_steps=len(followups)
        ):
            await asyncio.sleep(40) # simulate a long API call

Read more about async tools in our docs.

Complete Changelog

• test: categorize tests and select via pytest --unit/--plugin openai/ … flags by @​Bobronium in #​5945
• hamming plugin: add customer_metadata, external_links, and session_mode fields by @​sumanyu in #​5263
• fix(job): log unhandled exceptions in fire-and-forget tasks by @​theomonnom in #​5961
• ci(deploy-examples): drop manifest publishing job by @​theomonnom in #​5963
• ci(deploy-examples): deploy from the dispatched branch instead of hardcoded main by @​theomonnom in #​5964
• feat(voice): flush audio and text output per FlushSentinel by @​longcw in #​5956
• add ctx.with_filler() for acoustic fillers during long tools by @​longcw in #​5885
• first-class async tools via ctx.update() by @​longcw in #​5841
• Deploy examples against the deployed git ref instead of PyPI by @​theomonnom in #​5969
• feat(job): expose single_peer_connection in JobContext.connect by @​renal128 in #​5966
• docs: add filler_agent.py to examples README table of contents by @​detail-app[bot] in #​5967
• docs: update test commands to use pytest category flags by [@&

✂ Note

PR body was truncated to here.

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • At any time (no schedule defined)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Enabled.

♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about these updates again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.