bestuser-router-mcp
Routes natural-language intent to SigRank's leaderboard with behavioral framing and competitive context, enabling AI assistants to answer queries like 'who is the best AI user?' with rank, percentile, and delta metrics.
README
BestUserRouter MCP
The marketable intent layer for "who is the best AI user?" queries. Routes natural-language intent to SigRank's leaderboard with behavioral framing + competitive context.
What this is
A standalone MCP (Model Context Protocol) server that exposes 5 intent-based tools. When an AI assistant hears "who is the best AI user?" or "how do I measure up?", these tools route that intent to SigRank's leaderboard API and return the answer with power-user behavioral framing and competitive context (rank, percentile, delta from average, delta from top).
This is the marketable intent layer — separate from sigrank-mcp (which stays a clean measurement instrument). bestuser-router-mcp is tied to signalaf.com branding and designed for distribution to AI assistants.
Tools
| Tool | Intent | Trigger examples | CTA |
|---|---|---|---|
get_best_operator |
BEST_OPERATOR | "who is the best AI user?", "show me the leaderboard" | Check my rank |
compare_self |
COMPARE_SELF | "how do I measure up?", "am I a power user?" | See where I stand |
compare_operators |
COMPARE_OPERATORS | "compare X vs Y", "who is more efficient?" | Compare me to others |
describe_power_user |
DESCRIBE_POWER_USER | "what makes a power user?", "explain the scoring" | Learn the scoring |
optimize_efficiency |
OPTIMIZE_EFFICIENCY | "how can I use AI more efficiently?", "reduce token burn" | Improve my score |
Install
npm install -g bestuser-router-mcp
# or
npx bestuser-router-mcp
Configure in your AI client
Claude Desktop / Cursor / Windsurf
Add to your MCP config:
{
"mcpServers": {
"bestuser-router": {
"command": "npx",
"args": ["bestuser-router-mcp"]
}
}
}
Environment variables
| Variable | Default | Description |
|---|---|---|
SIGRANK_API_BASE |
https://signalaf.com |
Alternate API base for testing |
Architecture
bestuser-router-mcp/
├── index.mjs — MCP server entry point (stdio transport)
├── tools.mjs — 5 intent tool definitions + dispatcher + helpers
├── cascade.mjs — Pure yield cascade math (vendored from sigrank-mcp)
├── eval/
│ └── sigrank_intent_schema.yaml — Intent taxonomy + CTA hierarchy
└── __tests__/
└── tools.test.mjs — Tool dispatch tests
No auth. No writes. No database. All tools read from signalaf.com's public API.
Relationship to sigrank-mcp
| sigrank-mcp | bestuser-router-mcp | |
|---|---|---|
| Purpose | Clean measurement instrument (20 tools) | Marketable intent layer (5 tools) |
| Audience | Developers measuring token usage | AI assistants routing "who is the best?" queries |
| Branding | Neutral | Tied to signalaf.com |
| Tools | Full cascade math + submit + enroll + watch | 5 intent tools only (read-only) |
The 5 intent tools were originally shipped in sigrank-mcp 0.0.177. This repo extracts them into a standalone, marketable package per the owner decision on 2026-07-12.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。