Skip to main content

Real-time LiveKit agent visualizer — observe agent state, tool calls, and conversations without modifying your code.

Project description

Liveflow

liveflow

Real-time visualizer for LiveKit voice agents — see which agent is active, what tools are running, and follow conversations live. No changes to your agent code.

PyPI Python License GitHub


What is Liveflow?

Liveflow is a zero-instrumentation debugging companion for LiveKit Agents. It monkey-patches the LiveKit SDK's AgentSession to capture every event — state changes, tool calls, handoffs, transcripts — and streams them over a local WebSocket to the Liveflow VS Code extension, which renders a live dashboard with:

  • Agent Graph — all agents as nodes, active agent highlighted, animated handoff transitions
  • Tool Timeline — every @function_tool call with args, output, duration, and status
  • Conversation Transcript — live user/agent speech with interim transcripts
  • Chat Context Inspector — full LLM context window, system prompts, function calls, handoff markers

Install

pip install liveflow

Usage

Option 1 — VS Code Extension (recommended)

Install the Liveflow VS Code extension, open your agent project, and click Run with Liveflow in the editor title bar. The extension handles everything automatically — no terminal command needed.

Option 2 — Terminal

# Before
python agent.py dev

# After — drop-in replacement
liveflow agent.py dev

Your agent runs exactly as before. Liveflow captures everything transparently in the background and streams it to the VS Code dashboard.

How It Works

┌──────────────────────────┐   WebSocket    ┌──────────────────────┐
│  Liveflow Python Shim    │ ─────────────▶ │  VS Code Extension   │
│  • Patches AgentSession  │  JSON events   │  • ReactFlow graph   │
│  • Captures all events   │                │  • Tool timeline     │
│  • Local WS server       │                │  • Transcript view   │
└──────────────────────────┘                └──────────────────────┘
         ▲
         │  transparent monkey-patch
         │
┌──────────────────────────┐
│  Your agent.py           │  ← UNMODIFIED
│  (LiveKit Agents SDK)    │
└──────────────────────────┘

Source Code

github.com/21lakshh/Liveflow — contributions welcome!

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

liveflow-0.1.0.tar.gz (21.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

liveflow-0.1.0-py3-none-any.whl (25.7 kB view details)

Uploaded Python 3

File details

Details for the file liveflow-0.1.0.tar.gz.

File metadata

  • Download URL: liveflow-0.1.0.tar.gz
  • Upload date:
  • Size: 21.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for liveflow-0.1.0.tar.gz
Algorithm Hash digest
SHA256 6fa85f679d23b45fd1e69586a7ff0245af81934bb5030a60cd7f4475cd02f4b3
MD5 0701d48154e0b5e914a58a85a75b88dd
BLAKE2b-256 30ec44f499427c6261bb338725dcb0646bb59c978b920e73922525e03fd5f01a

See more details on using hashes here.

File details

Details for the file liveflow-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: liveflow-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 25.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for liveflow-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 9ec2a638ec0693f429d8264472466ae009c27469f885e2a83a006bb15d2be8ff
MD5 efbddf6c6b7aae9f2707c87d66834ec3
BLAKE2b-256 22fecb98dac396ce5f466e2c75aef01151fb40e9124391bc0094abcd1e0355aa

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page