livekit · u9g · Jul 31, 2026 · Jul 31, 2026
@@ -56,9 +56,9 @@ pytest with the `--allow-uncategorized` option to temporarily disable this rule
 
 ### Running Agents
 ```bash
-python myagent.py console   # Terminal mode with local audio I/O (no server needed)
-python myagent.py dev       # Development mode with hot reload (connects to LiveKit)
-python myagent.py start     # Production mode
+lk agent console myagent.py  # Terminal mode with local audio I/O (no server needed)
+lk agent dev myagent.py      # Development mode with hot reload (connects to LiveKit)
+python myagent.py start      # Production mode
 python myagent.py connect --room <room> --identity <id>  # Connect to existing room
 ```
 

@@ -351,16 +351,17 @@ For more examples and detailed setup instructions, see the [examples directory](
 ### Testing in terminal
 
 ```shell
-python myagent.py console
+lk agent console myagent.py
 ```
 
 Runs your agent in terminal mode, enabling local audio input and output for testing.
 This mode doesn't require external servers or dependencies and is useful for quickly validating behavior.
+It needs the [LiveKit CLI](https://docs.livekit.io/reference/developer-tools/livekit-cli/#setup).
 
 ### Developing with LiveKit clients
 
 ```shell
-python myagent.py dev
+lk agent dev myagent.py
 ```
 
 Starts the agent server and enables hot reloading when files change. This mode allows each process to host multiple concurrent agents efficiently.
@@ -402,7 +403,7 @@ uv sync --all-extras --dev
 This project includes many examples in the [`examples`](examples/) directory. To run them, create the file `examples/.env` with credentials for LiveKit Server and any necessary model providers (see `examples/.env.example`), then run:
 
 ```shell
-uv run examples/voice_agents/basic_agent.py dev
+lk agent dev examples/voice_agents/basic_agent.py
 ```
 
 For more information, see the [examples README](examples/README.md).

@@ -91,7 +91,7 @@ uv sync --all-extras --dev
 Run an example agent:
 
 ```bash
-uv run examples/voice_agents/basic_agent.py console
+lk agent console examples/voice_agents/basic_agent.py
 ```
 
 Your agent is now running in the console.

@@ -13,7 +13,7 @@ Try it in the [LiveKit Playground](https://agents.livekit.io/?example=avatar).
 > authenticates with your LiveKit credentials and the gateway creates the
 > provider session with LiveKit's wholesale key. Requires the
 > `avatar_lemonslice` feature flag on your project. Run with
-> `python inference_agent.py dev` and set `LIVEKIT_URL`, `LIVEKIT_API_KEY`,
+> `lk agent dev inference_agent.py` and set `LIVEKIT_URL`, `LIVEKIT_API_KEY`,
 > `LIVEKIT_API_SECRET`, and `LEMONSLICE_IMAGE_URL`.
 
 ## What's in here
@@ -44,7 +44,7 @@ Then:
 
 ```bash
 pip install -r requirements.txt
-python agent.py dev
+lk agent dev agent.py
 ```
 
 Connect from any LiveKit client. The agent reads the starting persona

@@ -16,7 +16,7 @@
 Env: LIVEKIT_URL, LIVEKIT_API_KEY, LIVEKIT_API_SECRET, LEMONSLICE_IMAGE_URL.
 
 Run:
-    python inference_agent.py dev
+    lk agent dev inference_agent.py
 """
 
 import logging

@@ -25,10 +25,10 @@ credentials in `../.env` or the environment:
 
 ```bash
 pip install -r requirements.txt
-python agent.py console
+lk agent console agent.py
 ```
 
-Use `python agent.py dev` to connect the agent to LiveKit Cloud for a frontend or
+Use `lk agent dev agent.py` to connect the agent to LiveKit Cloud for a frontend or
 telephony session.
 
 ## Tests and evals

@@ -9,9 +9,9 @@ the playground in real time.
 
 ```bash
 python examples/hotel_receptionist/fake_data/seed.py
-python examples/hotel_receptionist/agent.py console
+lk agent console examples/hotel_receptionist/agent.py
 # or, with the LiveKit playground:
-python examples/hotel_receptionist/agent.py dev
+lk agent dev examples/hotel_receptionist/agent.py
 ```
 
 `fake_data/seed.py` prints sample confirmation codes you can use to try

@@ -20,7 +20,7 @@ The agent registers one handler per control. All three (STT, LLM, TTS) call `upd
 
 ```bash
 pip install -r requirements.txt
-python agent.py dev
+lk agent dev agent.py
 ```
 
 The model list shown in the playground is sourced from `examples/playground.yaml`. To add or remove options, edit the `controls` block on the `inference` example there.
@@ -40,14 +40,14 @@ You need a LiveKit configured SIP trunk to allow the agents to receive inbound c
 
 2.  **Start the Mock Bank (The Target)**
     ```bash
-    uv run python examples/bank-ivr/ivr_system_agent.py dev
+    lk agent dev examples/bank-ivr/ivr_system_agent.py
     ```
     This agent acts as the IVR system waiting for calls.
 
 3.  **Start the Navigator Agent (The Caller)**
     Open a new terminal. This agent will wait for a dispatch job to tell it to call.
     ```bash
-    uv run python examples/bank-ivr/ivr_navigator_agent.py dev
+    lk agent dev examples/bank-ivr/ivr_navigator_agent.py
     ```
 
 4.  **Trigger the Call**

@@ -31,7 +31,7 @@ uv add "livekit-agents[xai,silero]" livekit-plugins-noise-cancellation
 You can talk to Grok directly in your terminal:
 
 ```bash
-uv run grok_voice_agent_api.py console
+lk agent console grok_voice_agent_api.py
 ```
 
 <img width="2012" height="626" alt="livekit-grok-voice-agents-api-console" src="https://github.com/user-attachments/assets/44f48804-a8c7-4c83-ade9-6ce9bec502ac" />
@@ -41,7 +41,7 @@ uv run grok_voice_agent_api.py console
 LiveKit hosts a playground environment where you can connect to the agent running on your machine. First run the agent:
 
 ```bash
-uv run grok_voice_agent_api.py dev
+lk agent dev grok_voice_agent_api.py
 ```
 
 Then: 
@@ -54,7 +54,7 @@ Then:
 LiveKit offers a bunch of [agent frontend](https://docs.livekit.io/frontends/) starter templates across languages. Like Option 2, the first step is to run the agent on your machine:
 
 ```bash
-uv run grok_voice_agent_api.py dev
+lk agent dev grok_voice_agent_api.py
 ```
 
 Then:

@@ -45,5 +45,5 @@ result = await WarmTransferTask(
 **Run the agent**
 
 ```python
-python warm_transfer.py dev
+lk agent dev warm_transfer.py
 ```