这是一款主要用于实验和探索LubeLogger中AI驱动功能的技术工具。

人工智能和MCP仍在不断发展的技术中，该项目的实施可能会在不提前通知的情况下中止。

先修条件

• 能够调用工具/外部集成的 AI 代理（例如：Claude Desktop）
• Node.js（代表NPX）

开始

该MCP服务器允许在AI代理级别（头部认证）或MCP服务器级别（本地认证）配置认证。

头部认证设置更复杂，但安全性更高;而本地认证更易设置，但允许用户无限制访问实例内所有车辆。

有关两种配置的详细信息，请参见维基百科

支持的指令

• 检索车辆列表
• 从车辆识别码（VIN）添加车辆
• 从收据图像添加燃料记录
• 从发票中添加服务/维修/升级记录
• 从仪表盘图像添加里程表记录
• 从发票图像添加供应（或店铺供应）记录
• 获取车辆最新的里程表读数
• LubeLogger实例的检查状态

示例用法

MCP Server for LubeLogger

This is a technical tool mainly for experimentation and exploration into AI-enabled features in LubeLogger.

AI and MCP are still evolving technologies, implementations of this project is subject to break without prior notice.

Pre-requisites

• AI agent with ability to call tools/external integrations(i.e.: Claude Desktop)
• Node.js(for npx)

Getting Started

This MCP Server allows for authentication to be configured at the AI-Agent level(Header Auth) or at the MCP Server level(Local Auth).

Header Auth is more complex to set up but offers greater security, whereas Local Auth is easier to set up but allows users unfettered access to all vehicles in the instance.

For details on both configurations, see Wiki

Commands Supported

• Retrieve list of vehicles
• Add Vehicle from VIN
• Add Fuel Record from image of receipt
• Add Service/Repair/Upgrade records from invoice
• Add Odometer record from image of dashboard
• Add Supply(or shop supply) record from image of invoice
• Get latest odometer reading from vehicle
• Check status of LubeLogger instance

Example Usage