Skip to main content
MCP server CX Codex

SigRank — Claude Code MCP server: what it does, how to install

Connects your coding assistant to SignalAF token telemetry to analyze cascade metrics, caching efficiency, and operator profiles.

Install into Claude Code · Platform CLI: remote server All methods and platforms ↓
claude mcp add --transport http sigrank https://signalaf.com/api/mcp
Open on GitHub

What SigRank — AI Operator Benchmarking does

This server connects your assistant directly to the SigRank benchmark hosted at https://signalaf.com/api/mcp. It inspects token flow across four primary pillars: input, output, cache-write, and cache-read. The Upsilon engine processes these raw figures to calculate token yield alongside metrics like Velocity and SNR.

Privacy is preserved by design. Telemetry snapshots do not collect source code, prompt contents, or model completions, sending only numerical counters and device signature metadata.

These numbers track token-handling efficiency and prompt cache utilization. They do not represent code quality.

What it helps with

  • inspect token processing ratios and cache utilization
  • pull operator telemetry benchmarks from the SignalAF leaderboard
  • calculate the Upsilon token yield metric for your setup
  • track context token patterns without sharing project source code

Example prompts for the agent

  • Check my operator stats and token metrics via the SigRank MCP server
  • Fetch the current leaderboard rankings and compare token yield values
  • Review my cache-read and cache-write telemetry on SignalAF

How to install SigRank — AI Operator Benchmarking

Platform CLI: remote server

claude mcp add --transport http sigrank https://signalaf.com/api/mcp

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about SigRank — AI Operator Benchmarking

How do I install SigRank in Claude Code?

Run claude mcp add --transport http sigrank https://signalaf.com/api/mcp in your terminal.

What does SigRank measure?

It tracks input, output, cache-write, and cache-read counters to determine token cascade patterns. The platform derives metrics such as Upsilon yield, SNR, and velocity from these raw counts.

Does SigRank upload proprietary code or prompt text?

No, ranked submissions exclude prompt text, model answers, and source files. The service processes only telemetry numbers, platform flags, and device signature data.

Similar MCP servers

All MCP servers for Agent setup: memory, context, orchestration →

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor