Krep MCP Server

Name: Krep MCP Server
Author: MCP-Mirror

@MCP-Mirror

代码仓库

关于 Krep MCP Server

Mirror of

#krep #mcp #string-search

基本信息

分类

其他

运行时

node

传输方式

stdio

发布者

MCP-Mirror

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "krep": {
      "command": "node",
      "args": [
        "/path/to/krep-mcp-server/src/index.js"
      ],
      "env": {
        "CLAUDE_MCP": "true",
        "KREP_PATH": "/path/to/krep-native/krep",
        "DEBUG": "true"
      },
      "description": "High-performance string search utility with unified interface",
      "disabled": false,
      "autoApprove": [
        "krep"
      ]
    }
  }
}

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 Krep MCP Server?

Krep MCP Server is a high-performance string search utility with MCP (Model Context Protocol) integration for the infinity-topos environment. It wraps the krep binary, an ultra-fast pattern matching utility, and exposes its functionality through MCP so AI assistants can perform efficient pattern searching in files and strings.

How to use Krep MCP Server?

First, ensure the krep binary is installed (e.g., make in the krep-native repository). Then configure the MCP server in your MCP settings file with the command, args, and env as shown in the README. Invoke the tool by calling the single unified function with parameters: pattern, target, mode (file, string, or count), caseSensitive, and threads. Integration with Cline VSCode extension is also supported via provided shell scripts.

Key features of Krep MCP Server

High-Performance Search using optimized algorithms (KMP, Boyer-Moore-Horspool, Rabin-Karp)
Hardware Acceleration: SIMD instructions (SSE4.2/AVX2/NEON) when available
Optimized Multi-Threading: automatically uses all CPU cores
Unified Interface: single function with file, string, and count modes
MCP Integration: seamless connection with AI assistants

Use cases of Krep MCP Server

Searching for patterns in large source code files or logs
Matching strings within application output or data streams
Counting occurrences of a pattern across files without extracting full matches
Enabling AI assistants to perform rapid codebase analysis
Integrating with Cline VSCode extension for in-editor pattern searching

FAQ from Krep MCP Server

What dependencies does Krep MCP Server require?

You need the krep binary installed and built from source (https://github.com/barton-willis/krep-native), Node.js to run the server, and an MCP‑compatible host (e.g., Claude Desktop, Cline). The server uses environment variables like KREP_PATH, CLAUDE_MCP, and DEBUG.

How do I configure the path to the krep binary?

Set the KREP_PATH environment variable in your MCP settings file to the full path of the krep binary. The server looks in 37 different paths, but it is recommended to specify it explicitly.

What search algorithms does Krep MCP Server use?

It automatically selects KMP for patterns shorter than 3 characters, Boyer-Moore-Horspool for 3–16 characters, and Rabin-Karp for patterns longer than 16 characters. SIMD instructions (SSE4.2, AVX2, NEON) are used when available.

Can I test the server without a real krep binary?

Yes. Set KREP_TEST_MODE to "true" to enable mock responses that simulate success even when the binary is missing. Also set KREP_SKIP_CHECK to "true" to skip binary existence checks.

How do I integrate Krep MCP Server with Cline VSCode extension?

Use the provided scripts: run install-cline-integration.sh to set up integration, run-cline-test.sh to test before installing, and uninstall-cline-integration.sh to remove it. After setup, use the command /krep krep pattern="..." target="..." mode="..." in Cline conversations.