Splitrail

What it does

Splitrail connects to your AI coding agents' local usage logs to provide a real-time dashboard of token consumption and spending. It allows AI assistants to query your usage statistics programmatically via MCP, helping you manage costs and understand model efficiency.

Tools

• get_daily_stats: Query usage statistics with date filtering
• get_model_usage: Analyze model usage distribution
• get_cost_breakdown: Get cost breakdown over a date range
• get_file_operations: Get file operation statistics
• compare_tools: Compare usage across different AI coding tools
• list_analyzers: List available analyzers

Installation

Run the following command to start the MCP server:

splitrail mcp

For Claude Desktop, add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "splitrail": {
      "command": "splitrail",
      "args": ["mcp"]
    }
  }
}

Supported hosts

Confirmed support for Claude Code, Gemini CLI, Qwen Code, Codex CLI, Cline, Roo Code, Kilo Code, GitHub Copilot, OpenCode, and Pi Agent.