MCP Server

Give AI coding assistants direct access to VibeCheck bug reports using the Model Context Protocol (MCP).

Overview

The VibeCheck MCP server (@vibecheck-mcp/mcp) lets AI coding assistants access your bug report data directly. Instead of copying and pasting console logs or network errors, your AI tool can fetch and analyze the full track data automatically.

MCP (Model Context Protocol) is an open standard that lets AI tools connect to external data sources. Once configured, you can simply paste a track URL and your AI assistant will pull in all the relevant debugging context.

Available tools

The MCP server provides five tools:

ToolDescription
get_trackFetch full track details — metadata, environment info, console logs, network requests, user actions, and performance metrics
analyze_track_errorsExtract and summarize all errors from console logs and failed network requests
get_track_networkGet detailed network request data — URLs, methods, status codes, timings, and response bodies
get_track_performanceGet Web Vitals and performance metrics (LCP, FID, CLS, TTFB, FCP)
get_track_actionsGet the timeline of user actions (clicks, scrolls, form inputs, navigation)

Setup

Claude Code

Run this command to add the MCP server:

claude mcp add vibecheck -- npx -y @vibecheck-mcp/mcp

Claude Desktop

Add to your Claude Desktop config file (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "vibecheck": {
      "command": "npx",
      "args": ["-y", "@vibecheck-mcp/mcp"]
    }
  }
}

Cursor

Add to your Cursor MCP settings (.cursor/mcp.json in your project or global config):

{
  "mcpServers": {
    "vibecheck": {
      "command": "npx",
      "args": ["-y", "@vibecheck-mcp/mcp"]
    }
  }
}

Windsurf

Add to your Windsurf MCP config (~/.codeium/windsurf/mcp_config.json):

{
  "mcpServers": {
    "vibecheck": {
      "command": "npx",
      "args": ["-y", "@vibecheck-mcp/mcp"]
    }
  }
}

VS Code

Add to your VS Code settings (.vscode/mcp.json):

{
  "servers": {
    "vibecheck": {
      "command": "npx",
      "args": ["-y", "@vibecheck-mcp/mcp"]
    }
  }
}

Usage examples

Once configured, you can reference VibeCheck tracks by URL in your AI assistant:

Fix a bug from a track

Fix this bug: https://app.vibecheck-qa.com/tracks/abc123

The AI will fetch the track data, analyze the console errors and network failures, and suggest a fix in the context of your codebase.

Analyze errors

Analyze errors in https://app.vibecheck-qa.com/tracks/abc123

Uses analyze_track_errors to extract all errors and failed requests from the track.

Debug network issues

Why is the API returning 500 errors in this track?
https://app.vibecheck-qa.com/tracks/abc123

Uses get_track_network to inspect request/response details.

Check performance

Are there any performance issues in this track?
https://app.vibecheck-qa.com/tracks/abc123

Uses get_track_performance to check Web Vitals against recommended thresholds.

Reproduce user steps

What steps did the user take before the error?
https://app.vibecheck-qa.com/tracks/abc123

Uses get_track_actions to show the full user action timeline.

Screenshots

The MCP server works with text-based track data (console logs, network requests, user actions, performance metrics). Screenshots and video recordings are not sent to the AI tools — only structured data that can be analyzed programmatically.

Requirements

  • Node.js 18 or later
  • npm (npx comes bundled with npm)
  • An AI tool that supports MCP (Claude Code, Claude Desktop, Cursor, Windsurf, VS Code)