减肥打卡记录

减肥进度追踪助手。用户告诉 AI 今天的体重，AI 自动帮您记录并生成漂亮的图表，清晰展示减肥进度。支持中英文界面，macOS/Windows/Linux 多平台使用。当用户说"记录体重"、"今天体重"、"减肥打卡"、"体重多少"时自动触发。

Security Scan

The skill is internally consistent with a local weight-tracker: it reads/writes a local JSON, serves an HTML page, and uses a small setup script — nothing requests unrelated credentials or remote installs.

Purpose & Capability

Name/description (weight tracking + chart) match the included files and required operations: config.json, weight_history.json, jianfei.html and a setup.sh to pick a port. No unrelated credentials, binaries, or services are requested.

Instruction Scope

SKILL.md instructs the agent to read/modify weight_history.json, start a local HTTP server, open the local page with the browser tool, screenshot it, then stop the server. Those actions are appropriate for generating the chart image, but they do involve running shell commands and manipulating local files — which is expected for a local web-based tracker.

Install Mechanism

No install spec is provided; this is instruction-only with bundled assets. The included setup.sh is present but contains only local logic (platform detection, port probing, interactive prompts) and does not download external resources. No remote archive downloads or 3rd-party package installs were found.

Credentials

The skill requests no environment variables, no credentials, and no config paths outside its own asset directory. Access to local JSON files and launching a local HTTP server are proportional to the stated functionality.

Persistence & Privilege

always:false and no special persistence or modification of other skills/systems. The skill runs interactively and does not request elevated persistent privileges.

Assessment

This skill appears to be what it says: a small local web app that stores weight records in weight_history.json and serves jianfei.html. Things to consider before installing: 1) The included setup.sh and the SKILL.md ask you to run shell commands — inspect setup.sh (you can see it) before running it. 2) The instructions start a Python HTTP server without an explicit bind host; by default that can be reachable from the local network. If you want it strictly local, run: python3 -m http.server <PORT> --bind 127.0.0.1. 3) The skill will read/write weight_history.json and config.json in the skill directory — back them up if they contain data you care about. 4) Stopping the server uses kill/netstat/fuser commands; those are normal but review them if you have sensitive long-running processes. If you want extra assurance, run the HTML locally from a sandboxed machine or inspect the full jianfei.html to verify it does not contact external endpoints (the provided files appear to only load local JSON).

Weight Tracker - 体重追踪器

可视化追踪减肥进度，支持图表展示、进度对比、数据导出、中英文双语、macOS/Windows/Linux 跨平台。

文件结构

weight-tracker/
├── SKILL.md
└── assets/
    ├── jianfei.html          # 主页面
    ├── config.json            # 用户配置
    ├── weight_history.json    # 体重数据
    └── setup.sh              # 跨平台初始化向导

快速开始

1. 初始化（仅首次需要）

cd <skill-directory>/assets
chmod +x setup.sh
./setup.sh

向导会自动检测平台（macOS/Windows/Linux），并引导配置：

• 语言选择（en/zh）
• 端口选择（自动检测可用端口）
• 减肥目标（起始/目标体重、日期周期）
• 初始体重记录

2. 启动服务

macOS:

cd <jianfei.html目录>
python3 -m http.server <PORT>

Windows (PowerShell):

cd <jianfei.html目录>
python -m http.server <PORT>

Linux:

cd <jianfei.html目录>
python3 -m http.server <PORT>

3. 打开页面

浏览器访问: http://localhost:<PORT>/jianfei.html

配置文件说明

config.json

{
  "title": "Weight Tracker",
  "language": "en",
  "port": 8080,
  "startWeight": 70,
  "targetWeight": 60,
  "startDate": "2026-01-01",
  "endDate": "2026-03-31",
  "startDateLabel": "01/01",
  "endDateLabel": "03/31"
}

关键字段：

• language: zh = 中文界面，en = 英文界面
• port: 本地服务器端口号
• startDate / endDate: 减肥周期（YYYY-MM-DD 格式）

weight_history.json

{
  "records": [
    {"date": "2026-03-29", "weight": 68.5}
  ]
}

切换语言

修改 config.json 中的 language 字段为 zh 或 en，然后刷新页面即可切换界面语言。

AI 工作流程

记录体重数据

1. 读取 weight_history.json
2. 解析 records 数组
3. 检查当天是否已有记录
4. 如果有，更新体重值；如果没有，添加新记录
5. 写回文件

打开浏览器并截图（跨平台）

1. 启动本地服务器

   cd <jianfei.html目录>
   
   # macOS
   python3 -m http.server <PORT>
   
   # Windows (PowerShell)
   python -m http.server <PORT>
   
   # Linux
   python3 -m http.server <PORT>
   

2. 打开浏览器

   • 启动 HTTP 服务器后等待 1-2 秒
   • 用 browser 工具打开 http://localhost:<PORT>/jianfei.html

3. 调整视口

   • 内容区尺寸：450 × 749 CSS像素
   • 视口调整为内容区的 110%：495 × 824

4. 截图

   • 使用 browser 工具的 screenshot 功能

5. 关闭浏览器并停止服务器 ⭐重要

   • 关闭浏览器标签页
   • 停止 HTTP 服务器：

   # macOS
   lsof -ti :<PORT> | xargs kill
   
   # Linux
   fuser -k <PORT>/tcp 2>/dev/null || kill $(lsof -ti:<PORT>) 2>/dev/null
   
   # Windows (PowerShell)
   netstat -ano | findstr :<PORT>
   taskkill /PID <PID> /F
   

快捷键

• 按 W 键：导出当前数据为 JSON 文件

关键实现细节

1. HTML 通过 fetch API 加载外部 JSON 文件
2. 必须通过 HTTP 服务器访问（file:// 不支持 fetch）
3. setup.sh 自动检测平台并适配命令
4. 端口默认为 8080，可在 config.json 中修改
5. 日期格式化兼容 macOS (date -j) 和 Linux (date -d)

目录规划建议

用户应将文件放在自己的目录下，例如：

~/Desktop/weight-tracker/
├── jianfei.html
├── config.json
├── weight_history.json
└── setup.sh