krep

High-performance string search MCP server with automatic CPU core scaling

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio bmorphism-krep-mcp-server node /path/to/krep-mcp-server/src/index.js \
  --env DEBUG="true" \
  --env KREP_PATH="/path/to/krep-native/krep" \
  --env CLAUDE_MCP="true"

How to use

The Krep MCP Server provides a unified interface to the underlying krep binary, a fast string search tool. Through the MCP framework you can perform file searches, string searches, or count-only operations using a single function exposed by the server. The server is designed to leverage multiple threading and optional hardware acceleration where available, delivering highly performant pattern matching for AI-assisted workflows. To use it, invoke the MCP tool with the krep server name and specify the desired tool (krep) along with a JSON payload that describes the pattern, target, mode (file, string, or count), case sensitivity, and thread preferences. This enables AI assistants to search large codebases or text efficiently without needing separate grep-like utilities for each task.

How to install

Prerequisites:

Node.js installed on the host running the MCP server
The krep-native binary built and available at a known path

Install steps:

Install dependencies and clone the repository containing the MCP server (or fetch the release package).
Build or obtain the krep-native binary and ensure it is accessible at the KREP_PATH you configure.
Ensure you have an MCP configuration file (e.g., mcp-config.json) and include a server entry for krep as described below.
Start the MCP server process (e.g., using run.sh or an equivalent launcher) and verify that the krep MCP endpoint is reachable.

Example commands:

Build krep-native (if needed): cd /path/to/krep-native make
Run the MCP server (assuming a launcher script): ./run.sh

Configuration example (mcp-config.json): { "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"] } } }

Additional notes

Notes and tips:

Ensure the KREP_PATH points to the compiled krep binary from krep-native.
The server uses CLAUDE_MCP to enable integration with Claude-based tools; adjust DEBUG as needed for troubleshooting.
If you upgrade krep-native, verify compatibility with the MCP server wrapper.
The MCP payload should include pattern, target, mode (file|string|count), and optional threads. If threads is null or omitted, the server will use all available cores.
Use the autoApprove setting to automatically enable the krep server within MCP configurations.

Related MCP Servers

zen

1.1k

Selfhosted notes app. Single golang binary, notes stored as markdown within SQLite, full-text search, very low resource usage

MCP -Deepseek_R1

A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3)

mcp-fhir

A Model Context Protocol implementation for FHIR

mcp

Inkdrop Model Context Protocol Server

mcp-appium-gestures

This is a Model Context Protocol (MCP) server providing resources and tools for Appium mobile gestures using Actions API..

dubco -npm

The (Unofficial) dubco-mcp-server enables AI assistants to manage Dub.co short links via the Model Context Protocol. It provides three MCP tools: create_link for generating new short URLs, update_link for modifying existing links, and delete_link for removing short links.