# LiveKit Claude Code MCP Evaluation

**Category:** Voice Infra
**Score:** 44 (D)

This is the agent-readable Devtool Arena evaluation page for LiveKit on Claude Code MCP.

## Summary

| Metric | Value |
|--------|-------|
| Status | completed |
| Overall score | 44 |
| Grade | D |
| Eval score | 33 |
| Discovery score | 71 |
| Cost | $0.22 |
| Runtime | 5m 7s |
| Tool calls | 9 |
| Errors | 1 |
| Tokens used | 368381 |
| Success | No |

## Agent-Readiness Checklist

| Signal | Evidence |
|--------|----------|
| Context7 | Not found |
| llms.txt | Not found |
| MCP server | Not found |
| Typed SDK | Not found |
| OpenAPI | Not found |
| Agent skills | Not found |
| CLI | Not found |

## Evaluation Prompt

````text

# Task
Using the LiveKit MCP, manage real-time voice infrastructure.

1. Discover the MCP tools relevant to rooms, sessions, or calls
2. List existing rooms or sessions
3. Create a new room/session called "*****-leaderboard-room"
4. List rooms/sessions again to verify the new one appears
5. If supported, generate a join token or print connection details
6. Print all output clearly labeled

Use ONLY the MCP tools for the operation.

You MUST use the LiveKit MCP server to accomplish this task.
Use the MCP tools for the actual product operations. Do NOT use curl, fetch, raw HTTP requests, CLI commands, or SDK calls as substitutes.

Use the MCP tools' help/descriptions to understand what each tool does.
If you need additional context, check the documentation: https://docs.livekit.io/mcp/

## Setup
Do NOT assume the LiveKit MCP server is already installed or already running in this sandbox.
Review the installation and authentication details below before starting the task.

### Installation
Transport: `http`
This MCP server uses `http` transport.
Claude Code should connect to: `https://docs.livekit.io/mcp/`
Do not assume a local MCP package is already installed; the task depends on the remote server being reachable.
Bootstrap notes: VERIFIED live: POST https://docs.livekit.io/mcp/ with an MCP 'initialize' request returns HTTP 200, content-type text/event-stream, and serverInfo {name: 'livekit-docs', version: '1.5.0'} — this is a real, reachable streamable-HTTP MCP endpoint (Claude Code transport value 'http' covers streamable HTTP). No mcp-session auth challenge, no bearer token needed. Registration command verified against local `claude mcp add --help`: syntax `claude mcp add --transport http <name> <url>` matches exactly what LiveKit's docs (fetched via docs.livekit.io/intro/coding-agents and the livekit-cli README) show: `claude mcp add --transport http livekit-docs https://docs.livekit.io/mcp`. IMPORTANT MISMATCH vs stored hints: the only MCP server LiveKit actually documents/publishes is the 'LiveKit Docs MCP server' — it exposes docs-browsing/search tools (get_docs_overview, get_pages, docs_search, code_search, get_changelog, get_pricing_info, submit_docs_feedback), NOT real-time voice-infrastructure management (no room/participant/session tools were found). I could not find any published LiveKit MCP server, package, or CLI subcommand that manages rooms/participants via MCP — searched GitHub, npm, PyPI, and the livekit-cli and livekit/livekit repos with no hits for an infra-management MCP server.

### Authentication
No authentication step is required for this MCP server.

## Execution
1. The harness has already written the launch configuration for `livekit`. Start with the task directly; do not rewrite agent settings, `.mcp-runtime.json`, or build a separate MCP client unless the connected server still fails after a normal retry.
2. You do not need to enumerate every MCP tool up front; inspect only the relevant tools you need. Agent-facing MCP tool names may be namespaced like `mcp__livekit__<tool_name>`.
3. Complete the task described above using MCP tools for the actual product operations.
4. If the server fails to start or authenticate, debug the documented setup and retry. Do not fall back to raw HTTP, SDK, CLI, or manual JSON-RPC/MCP calls as a substitute for exposed MCP tools.

Work in the /home/daytona/app directory.
If you use Bash during setup or debugging, limit it to installation/configuration work and then return to MCP tools for the task itself.
If there are errors, debug and fix them until the task runs successfully.

## Summary
When you are done, write a JSON summary file to /home/daytona/app/results.json with this structure:
{
  "operations": [
    {"step": 1, "description": "What you did", "mcp_tool": "tool_name", "success": true/false, "output_summary": "brief result"},
    ...
  ],
  "overall_success": true/false,
  "notes": ["any relevant notes about the execution"]
}

````

## Grader Results

| Check | Passed | Weight | Score | Details |
|-------|--------|--------|-------|---------|
| Task | No | — | — | — |
| Llms Txt | Yes | — | 10 | — |
| Tool Count | Yes | — | 10 | — |
| Auth Method | No | — | 0 | — |
| Official Mcp | Yes | — | 5 | — |
| Output Schema | No | — | 0 | — |
| Error Messages | — | — | 0 | mcp__livekit__code_search: Poor error message (generic or stack trace); mcp__livekit__get_changelog: Poor error message (generic or stack trace) |
| Tool Stability | Yes | — | 5 | — |
| Security Posture | — | — | 9 | — |
| Tool Api Mapping | — | — | 14 | — |
| Description Clarity | — | — | 14 | — |
| Tool Description Length | Yes | — | 4 | — |

## Run Artifacts

| Artifact | Value |
|----------|-------|
| Conversation turns | 11 |
| Tool call traces | 9 |
| Generated files | 1 |
| Exit code | — |
| Completed at | 2026-09-27T14:25:40.93583+00:00 |

### Generated Files

- /home/daytona/app/results.json

## Related Pages

- [Claude Code MCP leaderboard](/leaderboard/claudecode/mcp)
- [Compare Claude Code MCP companies](/leaderboard/claudecode/mcp/compare)
- [Agent Landscape](/leaderboard/discoverability)

Canonical URL: https://devtoolarena.com/claudecode/mcp/livekit