{
  "markdown": "# LiveKit Agent Skills\n\nReusable skills for AI coding agents building with [LiveKit](https://livekit.io). These skills provide procedural knowledge that helps agents build voice AI applications more effectively.\n\n## Available Skills\n\n| Skill | Description |\n|-------|-------------|\n| **livekit-agents** | Architectural guidance for building low-latency voice AI agents with LiveKit Agents SDK. Covers workflow design, handoffs, tasks, and mandatory testing practices. |\n| **livekit-simulations** | Generate targeted test scenarios for a LiveKit agent — locally, from the agent's own code plus what the user wants stress-tested — and run them as simulations. Bundles persona/goal/challenge libraries and a stdlib scenario builder. |\n\n## Installation\n\n```bash\nnpx skills add livekit/agent-skills\n```\n\nSkills activate automatically when agents detect relevant tasks (e.g., \"build a voice agent\", \"create a LiveKit agent\").\n\n## Skill Structure\n\nEach skill contains:\n- `SKILL.md` — Agent instructions with behavioral guidance\n- `references/` — Supporting documentation\n\n## Using with MCP\n\nThese skills are designed to work alongside the [LiveKit Docs MCP server](https://docs.livekit.io/mcp). The skills provide *behavioral guidance* (how to think about building voice agents) while MCP provides *factual information* (current API signatures, configuration options).\n\nInstall the MCP server for best results. Installation instructions for all supported coding agents are available at:\n\n**https://docs.livekit.io/intro/mcp-server/**\n\n## Design Philosophy\n\nThese skills follow the \"freeze forever\" principle—they encode behavioral guidance that remains valid regardless of API changes:\n\n- **Behavior over knowledge**: Skills teach *how to approach* problems, not API specifics\n- **MCP for facts**: All factual information (API signatures, config options) comes from live documentation\n- **Testing required**: Every agent implementation must include tests\n- **Latency-first**: Voice AI requires different architectural thinking than text-based agents\n\n## Contributing\n\nContributions welcome! When adding or modifying skills:\n\n1. Ensure content passes the \"freeze forever\" test—would it still be correct if never updated?\n2. Direct agents to MCP/docs for all API specifics\n3. Include trigger phrases in the skill description\n4. Keep skills under 500 lines for context efficiency\n\n## License\n\nMIT\n",
  "bytes": 2387,
  "sha": "6d94caa71a9d9be1e1ded9f903854d2b370b6b320f40209cb91b272c308264c7",
  "repo_slug": "livekit/agent-skills",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/skl_livekit_agent_skills_livekit_agents_4e242419/readme"
}