LLMDotfiles MCP Server
A Model Context Protocol server that scans and exposes AI-related dotfiles and configuration files to LLM agents, helping them understand project context and guidelines.
README
LLMDotfiles MCP Server
An MCP (Model Context Protocol) server for scanning AI-related dotfiles and configuration files, and expose them to the agent tool.
Overview
This MCP server provides access to a variety of dotfiles and configuration resources that help LLMs understand project context and guidelines. This is an experimental project that seek to detect and exposa all the different rules, guidelines, settings and configuration files the different agentic extensions make use of. This also includes the llm.txt proposal.
Installation
Requires NodeJS 18.x
Clone the project to any local folder you want.
Do a npm install or pnpm install to download the dependencies.
Use it as a project/workspace MCP server, this is an example config file:
{
"mcp": {
"LLMDotFiles": {
"command": "npm",
"args": [
"start"
],
"env": {
"ProjectPath": "<ABSOLUTE PROJECT PATH>",
"externalFolders": "<ABSOLUTE PATH TO EXTRA RULES>;<ANOTHER EXTRA PATH>"
},
"cwd": "<MCP SERVER INSTALLATION PATH>"
}
}
}
It can work as a global server, but since the project path is required, can only serve one project/workspace at a time.
Resources API
This server implements the MCP Resources API to provide access to dotfiles:
- List Resources: Automatically scans the project to find available dotfiles
- Read Resources: Access the content of specific dotfiles using their URIs
- Read ALL Resources: returns the content of all the files detected by the server, concatenated in a single string.
Resource URIs
Resources are available using the file:/// URI scheme:
file:///{projectPath}/{filename}
The bulk of all resources can be obtained using LLMfiles:/// URI scheme:
Files and Folders Scanned
Project Files
- llms.txt
- ai-instructions.md
- project-context.md
- .github/copilot-instructions.md
- .continuerules
- .clinerules
- .roorules
- .kilocoderules
Project Folders
- .vscode/rules
- .roo/rules
- .kilocode/rules
- .cursor/rules
- .continue/rules
- .windsurf/rules
External Folders
- ~/Documents/Cline/Rules
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。