← Back to Skills Marketplace
209
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install iwatch-swim-tracker
Description
当用户发送的图片为 Apple Watch 健身的泳池游泳记录,或用户消息包含「游泳」关键词并附带图片时触发。识别游泳训练截图,提取距离、配速、心率等结构化数据,保存训练记录并生成趋势分析报告。适用于游泳训练数据追踪、历史对比和个性化建议。
README (SKILL.md)
iWatch Swim Tracker
触发判断
满足以下任一条件时触发本技能:
- 当收到用户消息包含图片时,图片内容为 Apple Watch 健身 App 的泳池游泳训练记录截图。
- 当用户消息中包含「游泳」关键词并附带图片时。
数据提取
从截图中尽可能提取以下字段。能识别多少就提取多少,不确定的字段不要填。
必填字段
date: 训练日期 → 格式YYYY-MM-DDtime_range: 训练时段 → 格式HH:MM-HH:MMpool_length: 泳池长度(米)total_distance: 总距离(米)laps: 趟数strokes: 各泳姿距离 →{"freestyle": {"distance": N}, "breaststroke": {"distance": N}}- 自由泳=freestyle、蛙泳=breaststroke、仰泳=backstroke、蝶泳=butterfly、混合泳=medley、浮板=kickboard
duration: 体能训练时间 → 格式H:MM:SSduration_seconds: 时长转换为秒avg_pace: 平均配速 → 格式M:SS/100mavg_pace_seconds: 平均配速转换为秒/100mavg_heart_rate: 平均心率(次/分)
可选字段
active_calories: 动态千卡total_calories: 总千卡数effort_score: 耗能评分(1-10)effort_level: 耗能等级,如「适中」「困难」等session_number: 第 N 次游泳(如用户提供)auto_sets: 自动组合数据
数据校验
- 泳姿距离之和应等于总距离
- 趟数 × 泳池长度 ≈ 总距离
- 日期年份应为当前年份
- 不确定的值不要猜,宁可不填
处理流程
Step 1:提取数据
从图片中提取上述字段,组装为 JSON 对象。
Step 2:保存数据
python3 {baseDir}/scripts/extract_swim_data.py '\x3Cjson_data>'
同一天的数据会自动覆盖更新。
Step 3:查询历史
python3 {baseDir}/scripts/query_history.py --days 14 --weeks 4 --date \x3CYYYY-MM-DD>
回复格式
🏊 游泳训练记录
📅 \x3Cdate> \x3Ctime_range>
🏊 总距离: \x3Ctotal_distance>m(\x3Claps>趟 × \x3Cpool_length>m池)
⏱️ 时长: \x3Cduration>
⚡ 平均配速: \x3Cavg_pace>
💓 平均心率: \x3Cavg_heart_rate> 次/分
🔥 消耗: \x3Cactive_calories> 动态千卡 / \x3Ctotal_calories> 总千卡
泳姿分布:
- 自由泳: \x3Cfreestyle_distance>m
- 蛙泳: \x3Cbreaststroke_distance>m
- 浮板: \x3Ckickboard_distance>m
📊 AI 分析:
\x3C基于本次数据和历史趋势,给出 2-3 句个性化分析和建议>
📈 近期趋势:
\x3C与最近几次训练的关键指标对比>
Usage Guidance
This skill appears to do what it says: split screenshots, extract structured swim metrics, save them locally, and compute trends. Before installing, note two small items: (1) the image-splitting script requires Pillow (the skill mentions how to install it but the dependency isn't declared), so ensure your runtime has Pillow or can create a venv and install it; (2) extracted health data are stored as plain JSON under swim_data/ in the skill directory — if you care about privacy, plan where those files will live and whether you need encryption or cleanup. If you expect the vision extraction to run on a remote service, verify that no unintended personal data will be sent outside your environment; the included scripts themselves do not make network calls.
Capability Analysis
Type: OpenClaw Skill
Name: iwatch-swim-tracker
Version: 1.0.0
The skill is a legitimate tool for extracting and analyzing Apple Watch swimming workout data from screenshots. It uses Python scripts (extract_swim_data.py, query_history.py, and split_screenshot.py) to process vision-extracted JSON data, store it in a local directory structure, and generate performance trends. No evidence of data exfiltration, malicious command execution, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
Name/description (Apple Watch swim screenshot → extract/save/analyze) match the included scripts: split_screenshot.py (image segmentation), extract_swim_data.py (validate and save JSON), and query_history.py (trend/weekly analysis). Requested binary (python3) is appropriate for these scripts.
Instruction Scope
SKILL.md instructs using the included Python scripts and to feed JSON from vision extraction; it does not instruct reading unrelated system files or sending data to external endpoints. Note: SKILL.md refers to {baseDir} and to platform-provided vision extraction — those are implementation assumptions. Also, SKILL.md suggests installing Pillow for image splitting but does not declare that as a required dependency.
Install Mechanism
No install spec (instruction-only) and all code is local; there are no remote downloads, package installs, or execution of externally fetched code. The only runtime dependency is Python and optionally Pillow (local install via pip/venv as suggested).
Credentials
The skill requests no environment variables or credentials, which is appropriate. It writes user health data to local swim_data/ files within the skill directory — expected for a local tracker, but users should be aware that sensitive health data is stored on disk in plain JSON without encryption or access controls.
Persistence & Privilege
always:false and normal user-invocable behavior. The skill only creates/reads files under its swim_data directory and does not modify other skills or system-wide agent configuration.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install iwatch-swim-tracker - After installation, invoke the skill by name or use
/iwatch-swim-tracker - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
iWatch Swim Tracker 1.0.0 – Initial Release
- 新增 Apple Watch 泳池游泳训练截图智能识别与结构化数据提取。
- 支持自动保存训练记录,并校验字段完整性与合理性。
- 可按提取数据生成个性化趋势分析和训练建议。
- 实现 14 天/4 周历史记录对比功能。
- 支持多泳姿细分统计,生成标准化回复报告。
Metadata
Frequently Asked Questions
What is iwatch-swim-tracker?
当用户发送的图片为 Apple Watch 健身的泳池游泳记录,或用户消息包含「游泳」关键词并附带图片时触发。识别游泳训练截图,提取距离、配速、心率等结构化数据,保存训练记录并生成趋势分析报告。适用于游泳训练数据追踪、历史对比和个性化建议。 It is an AI Agent Skill for Claude Code / OpenClaw, with 209 downloads so far.
How do I install iwatch-swim-tracker?
Run "/install iwatch-swim-tracker" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is iwatch-swim-tracker free?
Yes, iwatch-swim-tracker is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does iwatch-swim-tracker support?
iwatch-swim-tracker is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created iwatch-swim-tracker?
It is built and maintained by zygzzp (@zygzzp); the current version is v1.0.0.
More Skills