mcp-grep
A grep server implementation that exposes grep functionality through the Model Context Protocol (MCP), supporting natural language prompts and common grep options such as case-insensitive matching, context lines, and recursive searching.
README
MCP-Grep
A grep server implementation that exposes grep functionality through the Model Context Protocol (MCP).
Installation
Installing via Smithery
To install Grep Server for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install @erniebrodeur/mcp-grep --client claude
Manual Installation
pip install mcp-grep
Usage
MCP-Grep runs as a server that can be used by MCP-compatible clients:
# Start the MCP-Grep server
mcp-grep-server
# Or use the MCP Inspector for interactive debugging and testing
mcp-grep-inspector
The server exposes the following MCP functionality:
- Resource:
grep://info- Returns information about the system grep binary - Tool:
grep- Searches for patterns in files using the system grep binary
Features
- Information about the system grep binary (path, version, supported features)
- Search for patterns in files using regular expressions
- Support for common grep options:
- Case-insensitive matching
- Context lines (before and after matches)
- Maximum match count
- Fixed string matching (non-regex)
- Recursive directory searching
- Natural language prompt understanding for easier use with LLMs
- Interactive debugging and testing through MCP Inspector
Example API Usage
Using the MCP Python client:
from mcp.client import MCPClient
# Connect to the MCP-Grep server
client = MCPClient()
# Get information about the grep binary
grep_info = client.get_resource("grep://info")
print(grep_info)
# Search for a pattern in files
result = client.use_tool("grep", {
"pattern": "search_pattern",
"paths": ["file.txt", "directory/"],
"ignore_case": True,
"recursive": True
})
print(result)
Natural Language Prompts
MCP-Grep understands natural language prompts, making it easier to use with LLMs. Examples:
# Basic search
Search for 'error' in log.txt
# Case-insensitive search
Find all instances of 'WARNING' regardless of case in system.log
# With context lines
Search for 'exception' in error.log and show 3 lines before and after each match
# Recursive search
Find all occurrences of 'deprecated' in the src directory and its subdirectories
# Fixed string search (non-regex)
Search for the exact string '.*' in config.js
# Limited results
Show me just the first 5 occurrences of 'TODO' in the project files
# Multiple options
Find 'password' case-insensitively in all .php files, show 2 lines of context, and limit to 10 results
MCP Inspector Integration
MCP-Grep includes an MCP Inspector integration for interactive debugging and testing:
# Start the MCP Inspector with MCP-Grep
mcp-grep-inspector
This opens a web-based UI where you can:
- Explore available resources and tools
- Test grep operations with different parameters
- View formatted results
- Debug issues with your grep queries
Development
# Clone the repository
git clone https://github.com/erniebrodeur/mcp-grep.git
cd mcp-grep
# Install in development mode
pip install -e ".[dev]"
# Run tests
pytest
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。