Natural Context Provider

Name: Natural Context Provider
Author: portel-dev

@portel-dev

关于 Natural Context Provider

One MCP to Rule Them All. NCP consolidates over 50 scattered MCP tools into two unified tools (Find & Run). Your AI goes from confused to confident with 87% fewer tokens, zero tool confusion, and 5x faster responses. Perfect for teams drowning in MCP tool sprawl.

#orchestration #vector-search #rag #token-saving #mcp-management #less-cognitive-load #reduce-clutter

基本信息

分类

其他

传输方式

stdio

发布者

portel-dev

提交者

Arul Kumaran

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "ncp": {
      "command": "npx",
      "args": [
        "@portel/ncp"
      ]
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Natural Context Provider?

Natural Context Provider (NCP) is a tool that aggregates multiple MCP servers into a single intelligent orchestrator, exposing only two tools (find and run) to the AI while handling routing, discovery, health monitoring, and semantic matching behind the scenes. It is designed for power users with multiple MCPs who want their AI to be focused, efficient, and cost-effective.

How to use Natural Context Provider?

Install globally via npm install -g @portel/ncp, then add "ncp": { "command": "ncp" } to your AI client configuration.

Key features of Natural Context Provider

Transforms N scattered MCP servers into one orchestrator.
AI sees only two tools: find and run.
Reduces token consumption by 47–83+% depending on setup size.
Handles routing, discovery, health monitoring, and semantic matching.
2-minute setup with a single npm install command.

Use cases of Natural Context Provider

Power users with multiple MCPs whose AI becomes indecisive or picks wrong tools.
Reducing token costs and session early termination caused by tool schema overload.
Maintaining existing MCP capabilities while improving AI focus and performance.

FAQ from Natural Context Provider

How does Natural Context Provider differ from using MCPs directly?

It eliminates the paradox of too many tools by exposing only two simple tools, reducing token consumption and decision fatigue for the AI while preserving all underlying tool capabilities.