Open source WebRTC and realtime AI infrastructure
LiveKit is an open source WebRTC and realtime AI infrastructure project whose skills help an AI agent build, operate, test, and debug voice and chat agents built with LiveKit Agents and LiveKit Cloud. Its skills cover concrete tasks such as structuring agent workflows and handoffs, shipping and rolling back versions, writing turn-level tests in pytest or Vitest, driving live conversations with the agent debugger, and running and writing simulation scenarios. The set stands out for covering the full agent lifecycle, from first build through production operation and regression testing.
Agent Skills
7 skills · 74 stars · MIT
Building Livekit Agents Builds voice and chat AI agents with LiveKit Agents and LiveKit Cloud.
Debugging Livekit Agents Drives a live multi-turn conversation with a LiveKit agent running locally, using `lk agent debug…
Operating Livekit Agents Deploys and operates a LiveKit agent in production: shipping a version to LiveKit Cloud and rolli…
Reading Livekit Docs Looks up current LiveKit facts (API signatures, CLI flags, config options, model and provider sup…
Running Livekit Simulations Runs LiveKit agent simulations and acts on the results.
Testing Livekit Agents Writes turn-level tests for a LiveKit agent in the user''s normal test suite: pytest (Python) or…
Writing Livekit Scenarios Creates and maintains the scenarios a LiveKit agent simulation runs, and wires the agent to consu…