mcpdirectory.dev
9,129 servers · updated 23 Sept

Sigrank MCP server

community Active last commit 17 d
{
  "mcpServers": {
    "signalaf-sigrank": {
      "type": "http",
      "url": "https://signalaf.com/api/mcp"
    }
  }
}
file .mcp.json (in your project root)version 1.0.0

What it does

Sigrank is an MCP server that exposes the SigRank leaderboard and Upsilon measurement engine for evaluating AI operator efficiency. It provides access to cascade metrics, operator rankings, profiles, and diagnostic tools for analyzing token-processing patterns. It is for developers integrating AI operator evaluation and efficiency benchmarking into their applications.

  • Query the SigRank leaderboard to retrieve ranked operators and their yield metrics.
  • Calculate Yield scores from token counts (cache read, output, input) to measure cascade efficiency.
  • Access operator profiles and diagnostic data to analyze token-processing patterns and efficiency.

Tools this server exposes

ToolDoes
calculate_yieldCalculate Yield metric from four token counts using the formula Υ = (cache_read × output) / input².
get_leaderboardRetrieve the SigRank leaderboard with operator rankings sorted by Yield.
get_operator_profileFetch an operator's profile including cascade layer, class, fingerprint, and historical metrics.
run_diagnosticsExecute diagnostic analysis on token cascade efficiency and signal compounding patterns.

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.