[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-70988":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":32,"readmeContent":33,"aiSummary":34,"trendingCount":16,"starSnapshotCount":16,"syncStatus":35,"lastSyncTime":36,"discoverSource":37},70988,"agents","livekit\u002Fagents","livekit","A framework for building realtime voice AI agents 🤖🎙️📹 ","https:\u002F\u002Fdocs.livekit.io\u002Fagents",null,"Python",10928,3219,99,219,0,57,173,491,171,45,"Apache License 2.0",false,"main",true,[5,27,28,29,30,31],"ai","openai","real-time","video","voice","2026-06-12 02:02:46","\u003C!--BEGIN_BANNER_IMAGE-->\n\n\u003Cpicture>\n  \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"\u002F.github\u002Fbanner_dark.png\">\n  \u003Csource media=\"(prefers-color-scheme: light)\" srcset=\"\u002F.github\u002Fbanner_light.png\">\n  \u003Cimg style=\"width:100%;\" alt=\"The LiveKit icon, the name of the repository and some sample code in the background.\" src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Flivekit\u002Fagents\u002Fmain\u002F.github\u002Fbanner_light.png\">\n\u003C\u002Fpicture>\n\n\u003C!--END_BANNER_IMAGE-->\n\u003Cbr \u002F>\n\n![PyPI - Version](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Flivekit-agents)\n[![PyPI Downloads](https:\u002F\u002Fstatic.pepy.tech\u002Fbadge\u002Flivekit-agents\u002Fmonth)](https:\u002F\u002Fpepy.tech\u002Fprojects\u002Flivekit-agents)\n[![Slack community](https:\u002F\u002Fimg.shields.io\u002Fendpoint?url=https%3A%2F%2Flivekit.io%2Fbadges%2Fslack)](https:\u002F\u002Flivekit.io\u002Fjoin-slack)\n[![Twitter Follow](https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Ffollow\u002Flivekit)](https:\u002F\u002Ftwitter.com\u002Flivekit)\n[![Ask DeepWiki for understanding the codebase](https:\u002F\u002Fdeepwiki.com\u002Fbadge.svg)](https:\u002F\u002Fdeepwiki.com\u002Flivekit\u002Fagents)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002Flivekit\u002Flivekit)](https:\u002F\u002Fgithub.com\u002Flivekit\u002Flivekit\u002Fblob\u002Fmaster\u002FLICENSE)\n\n\u003Cbr \u002F>\n\nLooking for the JS\u002FTS library? Check out [AgentsJS](https:\u002F\u002Fgithub.com\u002Flivekit\u002Fagents-js)\n\n## What is Agents?\n\n\u003C!--BEGIN_DESCRIPTION-->\n\nThe Agent Framework is designed for building realtime, programmable participants\nthat run on servers. Use it to create conversational, multi-modal voice\nagents that can see, hear, and understand.\n\n\u003C!--END_DESCRIPTION-->\n\n## Features\n\n- **Flexible integrations**: A comprehensive ecosystem to mix and match the right STT, LLM, TTS, and Realtime API to suit your use case.\n- **Integrated job scheduling**: Built-in task scheduling and distribution with [dispatch APIs](https:\u002F\u002Fdocs.livekit.io\u002Fagents\u002Fbuild\u002Fdispatch\u002F) to connect end users to agents.\n- **Extensive WebRTC clients**: Build client applications using LiveKit's open-source SDK ecosystem, supporting all major platforms.\n- **Telephony integration**: Works seamlessly with LiveKit's [telephony stack](https:\u002F\u002Fdocs.livekit.io\u002Fsip\u002F), allowing your agent to make calls to or receive calls from phones.\n- **Exchange data with clients**: Use [RPCs](https:\u002F\u002Fdocs.livekit.io\u002Fhome\u002Fclient\u002Fdata\u002Frpc\u002F) and other [Data APIs](https:\u002F\u002Fdocs.livekit.io\u002Fhome\u002Fclient\u002Fdata\u002F) to seamlessly exchange data with clients.\n- **Semantic turn detection**: Uses a transformer model to detect when a user is done with their turn, helps to reduce interruptions.\n- **MCP support**: Native support for MCP. Integrate tools provided by MCP servers with one loc.\n- **Builtin test framework**: Write tests and use judges to ensure your agent is performing as expected.\n- **Open-source**: Fully open-source, allowing you to run the entire stack on your own servers, including [LiveKit server](https:\u002F\u002Fgithub.com\u002Flivekit\u002Flivekit), one of the most widely used WebRTC media servers.\n\n## Installation\n\nTo install the core Agents library, along with plugins for popular model providers:\n\n```bash\npip install \"livekit-agents[openai,silero,deepgram,cartesia,turn-detector]\"\n```\n\n## Docs and guides\n\nDocumentation on the framework and how to use it can be found [here](https:\u002F\u002Fdocs.livekit.io\u002Fagents\u002F)\n\n### Building with AI coding agents\n\nIf you're using an AI coding assistant to build with LiveKit Agents, we recommend the following setup for the best results:\n\n1. **Install the [LiveKit Docs MCP server](https:\u002F\u002Fdocs.livekit.io\u002Fmcp)** — Gives your coding agent access to up-to-date LiveKit documentation, code search across LiveKit repositories, and working examples.\n\n2. **Install the [LiveKit Agent Skill](https:\u002F\u002Fgithub.com\u002Flivekit\u002Fagent-skills)** — Provides your coding agent with architectural guidance and best practices for building voice AI applications, including workflow design, handoffs, tasks, and testing patterns.\n\n   ```shell\n   npx skills add livekit\u002Fagent-skills --skill livekit-agents\n   ```\n\nThe Agent Skill works best alongside the MCP server: the skill teaches your agent *how to approach* building with LiveKit, while the MCP server provides the *current API details* to implement it correctly.\n\n## Core concepts\n\n- Agent: An LLM-based application with defined instructions.\n- AgentSession: A container for agents that manages interactions with end users.\n- entrypoint: The starting point for an interactive session, similar to a request handler in a web server.\n- AgentServer: The main process that coordinates job scheduling and launches agents for user sessions.\n\n## Usage\n\n### Simple voice agent\n\n---\n\n```python\nfrom livekit.agents import (\n    Agent,\n    AgentServer,\n    AgentSession,\n    JobContext,\n    RunContext,\n    cli,\n    function_tool,\n    inference,\n)\nfrom livekit.plugins import silero\n\n\n@function_tool\nasync def lookup_weather(\n    context: RunContext,\n    location: str,\n):\n    \"\"\"Used to look up weather information.\"\"\"\n\n    return {\"weather\": \"sunny\", \"temperature\": 70}\n\n\nserver = AgentServer()\n\n\n@server.rtc_session()\nasync def entrypoint(ctx: JobContext):\n    session = AgentSession(\n        vad=silero.VAD.load(),\n        # any combination of STT, LLM, TTS, or realtime API can be used\n        # this example shows LiveKit Inference, a unified API to access different models via LiveKit Cloud\n        # to use model provider keys directly, replace with the following:\n        # from livekit.plugins import deepgram, openai, cartesia\n        # stt=deepgram.STT(model=\"nova-3\"),\n        # llm=openai.LLM(model=\"gpt-4.1-mini\"),\n        # tts=cartesia.TTS(model=\"sonic-3\", voice=\"9626c31c-bec5-4cca-baa8-f8ba9e84c8bc\"),\n        stt=inference.STT(\"deepgram\u002Fnova-3\", language=\"multi\"),\n        llm=inference.LLM(\"openai\u002Fgpt-4.1-mini\"),\n        tts=inference.TTS(\"cartesia\u002Fsonic-3\", voice=\"9626c31c-bec5-4cca-baa8-f8ba9e84c8bc\"),\n    )\n\n    agent = Agent(\n        instructions=\"You are a friendly voice assistant built by LiveKit.\",\n        tools=[lookup_weather],\n    )\n\n    await session.start(agent=agent, room=ctx.room)\n    await session.generate_reply(instructions=\"greet the user and ask about their day\")\n\n\nif __name__ == \"__main__\":\n    cli.run_app(server)\n```\n\nYou'll need the following environment variables for this example:\n\n- LIVEKIT_URL\n- LIVEKIT_API_KEY\n- LIVEKIT_API_SECRET\n\n### Multi-agent handoff\n\n---\n\nThis code snippet is abbreviated. For the full example, see [multi_agent.py](examples\u002Fvoice_agents\u002Fmulti_agent.py)\n\n```python\n...\nclass IntroAgent(Agent):\n    def __init__(self) -> None:\n        super().__init__(\n            instructions=f\"You are a story teller. Your goal is to gather a few pieces of information from the user to make the story personalized and engaging.\"\n            \"Ask the user for their name and where they are from\"\n        )\n\n    async def on_enter(self):\n        self.session.generate_reply(instructions=\"greet the user and gather information\")\n\n    @function_tool\n    async def information_gathered(\n        self,\n        context: RunContext,\n        name: str,\n        location: str,\n    ):\n        \"\"\"Called when the user has provided the information needed to make the story personalized and engaging.\n\n        Args:\n            name: The name of the user\n            location: The location of the user\n        \"\"\"\n\n        context.userdata.name = name\n        context.userdata.location = location\n\n        story_agent = StoryAgent(name, location)\n        return story_agent, \"Let's start the story!\"\n\n\nclass StoryAgent(Agent):\n    def __init__(self, name: str, location: str) -> None:\n        super().__init__(\n            instructions=f\"You are a storyteller. Use the user's information in order to make the story personalized.\"\n            f\"The user's name is {name}, from {location}\"\n            # override the default model, switching to Realtime API from standard LLMs\n            llm=openai.realtime.RealtimeModel(voice=\"echo\"),\n            chat_ctx=chat_ctx,\n        )\n\n    async def on_enter(self):\n        self.session.generate_reply()\n\n\n@server.rtc_session()\nasync def entrypoint(ctx: JobContext):\n    userdata = StoryData()\n    session = AgentSession[StoryData](\n        vad=silero.VAD.load(),\n        stt=\"deepgram\u002Fnova-3\",\n        llm=\"openai\u002Fgpt-4.1-mini\",\n        tts=\"cartesia\u002Fsonic-3:9626c31c-bec5-4cca-baa8-f8ba9e84c8bc\",\n        userdata=userdata,\n    )\n\n    await session.start(\n        agent=IntroAgent(),\n        room=ctx.room,\n    )\n...\n```\n\n### Testing\n\nAutomated tests are essential for building reliable agents, especially with the non-deterministic behavior of LLMs. LiveKit Agents include native test integration to help you create dependable agents.\n\n```python\n@pytest.mark.asyncio\nasync def test_no_availability() -> None:\n    llm = google.LLM()\n    async AgentSession(llm=llm) as sess:\n        await sess.start(MyAgent())\n        result = await sess.run(\n            user_input=\"Hello, I need to place an order.\"\n        )\n        result.expect.skip_next_event_if(type=\"message\", role=\"assistant\")\n        result.expect.next_event().is_function_call(name=\"start_order\")\n        result.expect.next_event().is_function_call_output()\n        await (\n            result.expect.next_event()\n            .is_message(role=\"assistant\")\n            .judge(llm, intent=\"assistant should be asking the user what they would like\")\n        )\n\n```\n\n## Examples\n\nFor more examples and detailed setup instructions, see the [examples directory](examples\u002F). For even more examples, see the [python-agents-examples](https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fpython-agents-examples) repository.\n\n\u003Ctable>\n\u003Ctr>\n\u003Ctd width=\"50%\">\n\u003Ch3>🎙️ Starter Agent\u003C\u002Fh3>\n\u003Cp>A starter agent optimized for voice conversations.\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fvoice_agents\u002Fbasic_agent.py\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\n\u003Ch3>🔄 Multi-user push to talk\u003C\u002Fh3>\n\u003Cp>Responds to multiple users in the room via push-to-talk.\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fvoice_agents\u002Fpush_to_talk.py\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\">\n\u003Ch3>🎵 Background audio\u003C\u002Fh3>\n\u003Cp>Background ambient and thinking audio to improve realism.\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fvoice_agents\u002Fbackground_audio.py\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\n\u003Ch3>🛠️ Dynamic tool creation\u003C\u002Fh3>\n\u003Cp>Creating function tools dynamically.\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fvoice_agents\u002Fdynamic_tool_creation.py\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\">\n\u003Ch3>☎️ Outbound caller\u003C\u002Fh3>\n\u003Cp>Agent that makes outbound phone calls\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Foutbound-caller-python\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\n\u003Ch3>📋 Structured output\u003C\u002Fh3>\n\u003Cp>Using structured output from LLM to guide TTS tone.\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fvoice_agents\u002Fstructured_output.py\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\">\n\u003Ch3>🔌 MCP support\u003C\u002Fh3>\n\u003Cp>Use tools from MCP servers\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fvoice_agents\u002Fmcp\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\n\u003Ch3>💬 Text-only agent\u003C\u002Fh3>\n\u003Cp>Skip voice altogether and use the same code for text-only integrations\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fother\u002Ftext_only.py\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\">\n\u003Ch3>📝 Multi-user transcriber\u003C\u002Fh3>\n\u003Cp>Produce transcriptions from all users in the room\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fother\u002Ftranscription\u002Fmulti-user-transcriber.py\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\n\u003Ch3>🎥 Video avatars\u003C\u002Fh3>\n\u003Cp>Add an AI avatar with Tavus, Bithuman, LemonSlice, and more\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Favatar_agents\u002F\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003Ctr>\n\u003Ctd width=\"50%\">\n\u003Ch3>🍽️ Restaurant ordering and reservations\u003C\u002Fh3>\n\u003Cp>Full example of an agent that handles calls for a restaurant.\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"examples\u002Fvoice_agents\u002Frestaurant_agent.py\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\n\u003Ch3>👁️ Gemini Live vision\u003C\u002Fh3>\n\u003Cp>Full example (including iOS app) of Gemini Live agent that can see.\u003C\u002Fp>\n\u003Cp>\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fvision-demo\">Code\u003C\u002Fa>\n\u003C\u002Fp>\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\n\u003C\u002Ftable>\n\n## Running your agent\n\n### Testing in terminal\n\n```shell\npython myagent.py console\n```\n\nRuns your agent in terminal mode, enabling local audio input and output for testing.\nThis mode doesn't require external servers or dependencies and is useful for quickly validating behavior.\n\n### Developing with LiveKit clients\n\n```shell\npython myagent.py dev\n```\n\nStarts the agent server and enables hot reloading when files change. This mode allows each process to host multiple concurrent agents efficiently.\n\nThe agent connects to LiveKit Cloud or your self-hosted server. Set the following environment variables:\n- LIVEKIT_URL\n- LIVEKIT_API_KEY\n- LIVEKIT_API_SECRET\n\nYou can connect using any LiveKit client SDK or telephony integration.\nTo get started quickly, try the [Agents Playground](https:\u002F\u002Fagents-playground.livekit.io\u002F).\n\n### Running for production\n\n```shell\npython myagent.py start\n```\n\nRuns the agent with production-ready optimizations.\n\n## Contributing\n\nThe Agents framework is under active development in a rapidly evolving field. We welcome and appreciate contributions of any kind, be it feedback, bugfixes, features, new plugins and tools, or better documentation. You can file issues under this repo, open a PR, or chat with us in the [LiveKit community](https:\u002F\u002Fdocs.livekit.io\u002Fintro\u002Fcommunity\u002F).\n\n### Development setup\n\nThis project uses [uv](https:\u002F\u002Fdocs.astral.sh\u002Fuv\u002F) for package management. To install dependencies for development:\n\n```shell\nuv sync --all-extras --dev\n```\n\n### Examples\n\nThis project includes many examples in the [`examples`](examples\u002F) directory. To run them, create the file `examples\u002F.env` with credentials for LiveKit Server and any necessary model providers (see `examples\u002F.env.example`), then run:\n\n```shell\nuv run examples\u002Fvoice_agents\u002Fbasic_agent.py dev\n```\n\nFor more information, see the [examples README](examples\u002FREADME.md).\n\n### Tests\n\nUnit tests are in the `tests` directory and can be run with:\n\n```shell\nuv run pytest tests\u002Ftest_tools.py\n```\n\nIntegration tests for each plugin require various API credentials and run automatically in GitHub CI for PRs submitted by project maintainers. See the [tests workflow](.github\u002Fworkflows\u002Ftests.yml) for details.\n\n### Formatting\n\nThis project uses [ruff](https:\u002F\u002Fgithub.com\u002Fastral-sh\u002Fruff) for formatting and linting:\n\n```shell\nuv run ruff format\nuv run ruff check --fix\n```\n\n### Documentation\n\nTo generate docs locally with [pdoc](https:\u002F\u002Fgithub.com\u002Fpdoc3\u002Fpdoc):\n\n```shell\nuv sync --all-extras --group docs\nuv run --active pdoc --skip-errors --html --output-dir=docs livekit\n```\n\n\u003C!--BEGIN_REPO_NAV-->\n\u003Cbr\u002F>\u003Ctable>\n\u003Cthead>\u003Ctr>\u003Cth colspan=\"2\">LiveKit Ecosystem\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\n\u003Ctbody>\n\u003Ctr>\u003Ctd>Agents SDKs\u003C\u002Ftd>\u003Ctd>\u003Cb>Python\u003C\u002Fb> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fagents-js\">Node.js\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>LiveKit SDKs\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-js\">Browser\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-swift\">Swift\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-android\">Android\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-flutter\">Flutter\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-react-native\">React Native\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Frust-sdks\">Rust\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fnode-sdks\">Node.js\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fpython-sdks\">Python\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-unity\">Unity\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-unity-web\">Unity (WebGL)\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-esp32\">ESP32\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fclient-sdk-cpp\">C++\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Starter Apps\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fagent-starter-python\">Python Agent\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fagent-starter-node\">TypeScript Agent\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fagent-starter-react\">React App\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fagent-starter-swift\">SwiftUI App\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fagent-starter-android\">Android App\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fagent-starter-flutter\">Flutter App\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fagent-starter-react-native\">React Native App\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit-examples\u002Fagent-starter-embed\">Web Embed\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>UI Components\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fcomponents-js\">React\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fcomponents-android\">Android Compose\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fcomponents-swift\">SwiftUI\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fcomponents-flutter\">Flutter\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Server APIs\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fnode-sdks\">Node.js\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fserver-sdk-go\">Golang\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fserver-sdk-ruby\">Ruby\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fserver-sdk-kotlin\">Java\u002FKotlin\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fpython-sdks\">Python\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Frust-sdks\">Rust\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fagence104\u002Flivekit-server-sdk-php\">PHP (community)\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FpabloFuente\u002Flivekit-server-sdk-dotnet\">.NET (community)\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Resources\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fdocs.livekit.io\">Docs\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fdocs.livekit.io\u002Fmcp\">Docs MCP Server\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Flivekit-cli\">CLI\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fcloud.livekit.io\">LiveKit Cloud\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>LiveKit Server OSS\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Flivekit\">LiveKit server\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fegress\">Egress\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fingress\">Ingress\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flivekit\u002Fsip\">SIP\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003C\u002Ftr>\n\u003Ctr>\u003Ctd>Community\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fcommunity.livekit.io\">Developer Community\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Flivekit.io\u002Fjoin-slack\">Slack\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fx.com\u002Flivekit\">X\u003C\u002Fa> · \u003Ca href=\"https:\u002F\u002Fwww.youtube.com\u002F@livekit_io\">YouTube\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\n\u003C\u002Ftbody>\n\u003C\u002Ftable>\n\u003C!--END_REPO_NAV-->\n","livekit\u002Fagents 是一个用于构建实时语音AI代理的框架。它支持创建能够看、听和理解的多模态对话代理，具备灵活集成多种语音转文字（STT）、大语言模型（LLM）、文本转语音（TTS）及实时API的能力，并内置了任务调度与分发机制，使得连接最终用户与代理变得简单高效。此外，该框架还提供了广泛的WebRTC客户端支持，可以无缝对接LiveKit的电话堆栈，实现与电话系统的互联互通；同时，通过RPC等数据交换接口，保证了与客户端之间的流畅交互。适用于需要高度定制化、可扩展性的客户服务、智能助手或虚拟接待员等场景。",2,"2026-06-11 03:35:18","high_star"]