← 返回 Skills 市场
141
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install coros-skill
功能描述
COROS 高驰跑步数据获取(跑步专项): - 自动登录 COROS 账号(支持 Token/Cookie 缓存) - 获取 Dashboard(训练状态、负荷分析、最近运动) - 获取活动列表与活动详情(含分圈、天气、训练效果) - 获取训练日程与训练目标汇总 触发词示例: - "查看高驰跑步数据" - "我的...
使用说明 (SKILL.md)
COROS 高驰跑步数据获取 Skill
基于 COROS API 的跑步数据读取与展示工具,面向日常训练复盘。
使用范围
- 当前环境仅使用
coros这个运动 Skill。 - 运动数据相关需求统一走本 Skill(高驰/COROS 跑步数据)。
功能概览
1. 智能登录
- 支持
account + p1 + p2登录 - 支持 Token/Cookie 缓存,减少重复登录
- 自动保存和恢复登录状态
2. Dashboard 数据
- 训练状态:短期负荷(ATI)、长期负荷(CTI)、负荷比
- 最近运动记录
- 心率数据统计
- 运动类型统计
- 本周训练汇总
3. 活动列表
- 完整的跑步历史记录
- 显示:日期、名称、距离、时长、配速、心率、训练负荷(TL)
- 支持分页查询
4. 活动详情
- 通过
activity/detail/query获取单次活动完整详情 - 展示模块:计圈数据、天气、概要数据、训练效果、运动感受
- 时间字段统一按
HH:MM:SS展示,便于和网页详情页对齐 - 当前默认展示“活动列表第一页第一条”的详情(通常是最近一次)
5. 训练日程
- 训练日程查询
- 训练目标汇总
- 支持训练日程新增/更新/删除(
training/schedule/update) - 内置安全模式:默认仅预览,确认后可开启真实写入
运行方式
OpenClaw 对话触发
示例(可直接复制):
查看我的高驰跑步数据我的跑步记录高驰运动分析最近跑步怎么样今天跑步了吗看最近一次跑步详情帮我复盘今天这次跑步
命令行运行
python3 main.py
配置文件
在 config.json 中配置账号与登录上下文:
{
"coros": {
"account": "[email protected]",
"p1": "$2b$10$xxx",
"p2": "$2b$10$xxx"
},
"cookie": "_c_WBKFRo=xxx; _nb_ioWEgULi=",
"demo_mode": false
}
发布到 clawhub.ai 时:
- 使用
.clawhubignore排除config.json、api.md、.claude/ - 将
config.example.json作为公开示例配置 - 本地仍保留
config.json以保证日常运行不受影响
API 端点
| 功能 | API |
|---|---|
| 登录 | POST /account/login |
| Dashboard详情 | GET /dashboard/detail/query |
| Dashboard汇总 | GET /dashboard/query |
| 周期纪录 | GET /dashboard/queryCycleRecord |
| 活动列表 | GET /activity/query |
| 活动详情 | POST /activity/detail/query |
| 训练日程 | GET /training/schedule/query |
| 日程汇总 | GET /training/schedule/querysum |
| 估算训练 | POST /training/program/estimate |
| 计算训练 | POST /training/program/calculate |
| 新增/更新/删除日程 | POST /training/schedule/update |
| 训练计划列表 | GET /training/program/list |
安全使用建议
This skill appears coherent for reading COROS data, but take these precautions before installing: 1) Keep config.json private — it contains account identifiers, password hashes (p1/p2) and cookies; do not upload it to public registries. 2) By default schedule writes are preview-only, but the skill can perform real writes if you enable auto_apply or change schedule_write settings — review those settings before enabling. 3) Verify network calls go to the expected COROS API domain (teamcnapi.coros.com) and run the code in an isolated environment if you don't trust the source. 4) After testing, rotate credentials if you suspect they were exposed. If you want higher assurance, inspect the full main.py runpath locally (it saves tokens to config.json) or run with demo_mode=true.
功能分析
Type: OpenClaw Skill
Name: coros-skill
Version: 0.1.0
The skill is a legitimate integration for the COROS fitness platform, designed to fetch and manage running activity data and training schedules. It interacts exclusively with the official COROS API (teamcnapi.coros.com) using standard Python libraries and handles authentication by storing session tokens and cookies in a local config.json file. There is no evidence of data exfiltration, malicious execution, or prompt injection intended to hijack the agent.
能力评估
Purpose & Capability
Name/description match behavior: the skill logs into COROS (using account/p1/p2 or cookies), reads dashboard, activities, activity details and training schedules, and exposes schedule write operations. The required artifacts (config.json with COROS account hashes and cookies) are proportional to these capabilities.
Instruction Scope
SKILL.md and main.py focus on COROS API calls and local config. The skill caches tokens/cookies to config.json and supports schedule write operations; by default preview/auto_apply=false is used, but the code supports making writes if configured. Users should be aware the skill reads/writes a local config file that contains sensitive data and controls whether writes occur.
Install Mechanism
No install spec (instruction-only with an included Python script). This has low installation risk — nothing is downloaded from arbitrary URLs or installed automatically.
Credentials
The skill does not request unrelated environment variables or external credentials; it expects credentials stored in a local config.json (account, p1/p2 hashes, cookies). Storing secrets in config.json is functional but requires care to avoid accidental exposure.
Persistence & Privilege
always:false and agent-autonomy defaults are normal. The skill persists its own token/cookie/state by writing to config.json (saving token, user_id, saved_cookie). This is limited to its own config file and does not modify other skills or global agent config.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install coros-skill - 安装完成后,直接呼叫该 Skill 的名称或使用
/coros-skill触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v0.1.0
Initial release of the coros-skill for retrieving and analyzing COROS running data.
- Supports automatic login with token/cookie caching.
- Fetches dashboard data: training status, recent activities, and weekly summary.
- Retrieves activity list and detailed training records, including lap, weather, and performance metrics.
- Allows viewing and previewing training schedules and goals.
- Simple chat trigger phrases and command line usage available.
元数据
常见问题
🏃 Coros 是什么?
COROS 高驰跑步数据获取(跑步专项): - 自动登录 COROS 账号(支持 Token/Cookie 缓存) - 获取 Dashboard(训练状态、负荷分析、最近运动) - 获取活动列表与活动详情(含分圈、天气、训练效果) - 获取训练日程与训练目标汇总 触发词示例: - "查看高驰跑步数据" - "我的... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 141 次。
如何安装 🏃 Coros?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install coros-skill」即可一键安装,无需额外配置。
🏃 Coros 是免费的吗?
是的,🏃 Coros 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
🏃 Coros 支持哪些平台?
🏃 Coros 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 🏃 Coros?
由 doaspx(@doaspx)开发并维护,当前版本 v0.1.0。
推荐 Skills