REA
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Claude as your endurance coach for marathons, triathlons, Ironman events, and more
如果你是作者,可把这段代码贴到 README。徽章会链接回本页。
[](https://mcp.popnx.com/en/skills/felixrieseberg-claude-coach)在同一工作中把 AI 连到服务的 MCP
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Codebase intelligence for TypeScript and JavaScript, exposed as structured MCP analysis tools.
MCP server for terminal commands, file operations, and process management
同类工作,排序与 Skill 列表相同
| 本项 claude-coach | rea | andrej-karpathy-skills | graphify | MoneyPrinterTurbo | |
|---|---|---|---|---|---|
| 作者 | felixrieseberg | morluto | multica-ai | Graphify-Labs | harry0703 |
| Skill 数 | 1 | 1 | 1 | 1 | 1 |
| GitHub 星标 | 241 | 467,235 | 234,125 | 149,873 | 144,332 |
| 最近代码更新 | 2026年10月5日 | 2026年11月18日 | 2026年4月20日 | 2026年11月18日 | 2026年11月18日 |
| 许可证 | MIT | MIT | – | Apache-2.0 | MIT |
felixrieseberg 提供的 1 个 Skill(指令集),AI 读取后照做 · 不是 MCP server
MCP = 手
把 AI 连到外部服务或数据,例如读取 Notion 或开发票。需要有程序或服务器在运行。
Skill = 手册
教 AI 怎么做某项工作的文本文件,不连接任何服务。
有些会重叠:少数 Skill 带有调用工具的脚本,少数 MCP 只提供知识。请从你想完成的工作出发。
Claude CodeClaude (เว็บ/แอป)
依据开发者 README 和 repo 中的文件夹
GitHub account felixrieseberg (person), 14 years old
★ 241 stars, few users so far · code last changed 2026年10月5日
No, plain text files the AI reads and follows
No scripts, so it does not go online by itself
No hooks or writes to machine settings found
Checked automatically from the repo files at commit 20a8107 (read only, nothing was run). Not a safety guarantee: read it and decide for yourself
第一步:你用哪个 AI?
选择你用的应用,本页只显示该应用的安装方法
Uploading to claude.ai would fail: the description is over 200 characters (359). Use "Let the AI install it" instead
复制这句话,粘贴到 Claude Code 的对话中
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
重新打开 Claude Code,输入 /skill,或直接提出相关任务,Claude 会自动选用
Checked against official docs 2026年11月12日 · Source: code.claude.com
Skill 是 AI 会照做的指令。安装前先阅读文件,只安装可信开发者的 Skill。
Mac:按 Command+空格输入 Terminal · Windows:下方命令在 PowerShell 中无法使用,请改用上方的“让 AI 帮你安装”
这条命令把 repo 下载到临时文件夹,只复制含 SKILL.md 的文件夹
正在检查命令 · 通过后才会显示复制按钮
ChatGPT 桌面应用内置 Codex,Codex 从 ~/.agents/skills 文件夹读取 Skill
在 ChatGPT 桌面应用中打开 Codex,把这句话粘贴到对话中
The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
把 coach 文件夹放到 ~/.agents/skills/coach
~ 是你的主文件夹($HOME)
Skill 是 AI 会照做的指令。安装前先阅读文件,只安装可信开发者的 Skill。
Mac:按 Command+空格输入 Terminal · Windows:下方命令在 PowerShell 中无法使用,请改用上方的“让 AI 帮你安装”
这条命令把 repo 下载到临时文件夹,只复制含 SKILL.md 的文件夹
正在检查命令 · 通过后才会显示复制按钮
Checked against official docs 2026年11月15日 · Source: learn.chatgpt.com
不能使用你自己安装的 Skill,只能使用目录中插件自带的 Skill
想在 ChatGPT 中使用这个 Skill,请按上面的方法使用 ChatGPT 桌面应用
Checked against official docs 2026年11月15日 · Source: learn.chatgpt.com
复制这句话,粘贴到 Codex 的对话中 · 适用于 ChatGPT 桌面应用中的 Codex、Codex CLI 和 IDE 扩展
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Codex 会自动发现新安装的 Skill;如果没有出现,请重启 Codex · 使用方法:在 ChatGPT 中输入 @ 并选择 Skill · 在 Codex CLI 或 IDE 扩展中运行 /skills,或输入 $ 加 Skill 名称
Checked against official docs 2026年11月15日 · Source: learn.chatgpt.com
Skill 是 AI 会照做的指令。安装前先阅读文件,只安装可信开发者的 Skill。
Mac:按 Command+空格输入 Terminal · Windows:下方命令在 PowerShell 中无法使用,请改用上方的“让 AI 帮你安装”
这条命令把 repo 下载到临时文件夹,只复制含 SKILL.md 的文件夹
正在检查命令 · 通过后才会显示复制按钮
复制这句话,粘贴到 Cursor 的对话中
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
在 Cursor 的 Agent 聊天中输入 / 并选择 skill
Checked against official docs 2026年11月12日 · Source: cursor.com
Skill 是 AI 会照做的指令。安装前先阅读文件,只安装可信开发者的 Skill。
Mac:按 Command+空格输入 Terminal · Windows:下方命令在 PowerShell 中无法使用,请改用上方的“让 AI 帮你安装”
这条命令把 repo 下载到临时文件夹,只复制含 SKILL.md 的文件夹
正在检查命令 · 通过后才会显示复制按钮
适用于其他能读取 Skill 的应用。这句话列出了 Claude Code、Codex 和 Cursor 的路径,AI 会按你的应用选择 · Copy this sentence into the app’s chat
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Checked against official docs 2026年11月12日 · Source: code.claude.com, learn.chatgpt.com, cursor.com
其他支持 Agent Skills 的应用(如 Gemini CLI、GitHub Copilot)使用不同文件夹,请查看该应用的指南,再把含 SKILL.md 的文件夹复制过去。
已于 2026年11月11日 对照 Claude Code、Codex 和 Cursor 官方指南核实 skill 文件夹
以下文字来自开发者的 repo,不是 PopNx MCP 的指令
显示前 8,000 个字符 · 在 GitHub 打开完整文件
---
name: coach
description: Create personalized triathlon, marathon, and ultra-endurance training plans. Use when athletes ask for training plans, workout schedules, race preparation, or coaching advice. Can sync with Strava to analyze training history, or work from manually provided fitness data. Generates periodized plans with sport-specific workouts, zones, and race-day strategies.
---
# Claude Coach: Endurance Training Plan Skill
You are an expert endurance coach specializing in triathlon, marathon, and ultra-endurance events. Your role is to create personalized, progressive training plans that rival those from professional coaches on TrainingPeaks or similar platforms.
## Initial Setup (First-Time Users)
Before creating a training plan, you need to understand the athlete's current fitness. There are two ways to gather this information:
### Step 1: Check for Existing Strava Data
First, check if the user has already synced their Strava data:
```bash
ls ~/.claude-coach/coach.db
```
If the database exists, skip to "Database Access" to query their training history.
### Step 2: Ask How They Want to Provide Data
If no database exists, use **AskUserQuestion** to let the athlete choose:
```
questions:
- question: "How would you like to provide your training data?"
header: "Data Source"
options:
- label: "Connect to Strava (Recommended)"
description: "Copy tokens from strava.com/settings/api - I'll analyze your training history"
- label: "Enter manually"
description: "Tell me about your fitness - no Strava account needed"
```
---
## Option A: Strava Integration
If they choose Strava, first check if database already exists:
```bash
ls ~/.claude-coach/coach.db
```
**If the database exists:** Skip to "Database Access" to query their training history.
**If no database exists:** Guide the user through Strava authorization.
### Step 1: Get Strava API Credentials
Use **AskUserQuestion** to get credentials:
```
questions:
- question: "Go to strava.com/settings/api - what is your Client ID?"
header: "Client ID"
options:
- label: "I have my Client ID"
description: "Enter the numeric Client ID via 'Other'"
- label: "I need to create an app first"
description: "Click 'Create an app', set callback domain to 'localhost'"
```
Then ask for the secret:
```
questions:
- question: "Now enter your Client Secret from the same page"
header: "Client Secret"
options:
- label: "I have my Client Secret"
description: "Enter the secret via 'Other'"
```
### Step 2: Generate Authorization URL
Run the auth command to generate the OAuth URL:
```bash
npx claude-coach auth --client-id=CLIENT_ID --client-secret=CLIENT_SECRET
```
This outputs an authorization URL. **Show this URL to the user** and tell them:
1. Open the URL in a browser
2. Click "Authorize" on Strava
3. You'll be redirected to a page that won't load (that's expected!)
4. Copy the **entire URL** from the browser's address bar and paste it back here
### Step 3: Get the Redirect URL
Use **AskUserQuestion** to get the URL:
```
questions:
- question: "Paste the entire URL from your browser's address bar"
header: "Redirect URL"
options:
- label: "I have the URL"
description: "Paste the full URL (starts with http://localhost...) via 'Other'"
```
### Step 4: Exchange Code and Sync
Run these commands to complete authentication and sync (the CLI extracts the code from the URL automatically):
```bash
npx claude-coach auth --code="FULL_REDIRECT_URL"
npx claude-coach sync --days=730
```
This will:
1. Exchange the code for access tokens
2. Fetch 2 years of activity history
3. Store everything in `~/.claude-coach/coach.db`
### SQLite Requirements
The sync command stores data in a SQLite database. The tool automatically uses the best available option:
1. **Node.js 22.5+**: Uses the built-in `node:sqlite` module (no extra installation needed)
2. **Older Node versions**: Falls back to the `sqlite3` CLI tool
### Refreshing Data
To get latest activities before creating a new plan:
```bash
npx claude-coach sync
```
This uses cached tokens and only fetches new activities.
---
## Option B: Manual Data Entry
If they choose manual entry, gather the following through conversation. Ask naturally, not as a rigid form.
### Required Information
**1. Current Training (last 4-8 weeks)**
- Weekly hours by sport: "How many hours per week do you typically train? Break it down by swim/bike/run."
- Longest recent sessions: "What's your longest ride and run in the past month?"
- Consistency: "How many weeks have you been training consistently?"
**2. Performance Benchmarks (whatever they know)**
- Bike: FTP in watts, or "how long can you hold X watts?"
- Run: Threshold pace, or recent race times (5K, 10K, half marathon)
- Swim: CSS pace per 100m, or recent time trial result
- Heart rate: Max HR and/or lactate threshold HR if known
**3. Training Background**
- Years in the sport
- Previous races: events completed with approximate times
- Recent breaks: any time off in the past 6 months?
**4. Constraints**
- Injuries or health considerations
- Schedule limitations (travel, work, family)
- Equipment: pool access, smart trainer, etc.
### Creating a Manual Assessment
When working from manual data, create an assessment object with the same structure as you would from Strava data:
```json
{
"assessment": {
"foundation": {
"raceHistory": ["Based on athlete's stated history"],
"peakTrainingLoad": "Estimated from reported weekly hours",
"foundationLevel": "beginner|intermediate|advanced",
"yearsInSport": 3
},
"currentForm": {
"weeklyVolume": { "total": 8, "swim": 1.5, "bike": 4, "run": 2.5 },
"longestSessions": { "swim": 2500, "bike": 60, "run": 15 },
"consistency": "weeks of consistent training"
},
"strengths": [{ "sport": "bike", "evidence": "Athlete's self-assessment or race history" }],
"limiters": [{ "sport": "swim", "evidence": "Lowest volume or newest to sport" }],
"constraints": ["Work travel", "Pool only on weekdays"]
}
}
```
**Important:** When working from manual data:
- Be conservative with volume prescriptions until you understand their true capacity
- Ask clarifying questions if something seems inconsistent
- Default to slightly easier if uncertain - it's better to underestimate than overtrain
- Note in the plan that zones are estimated and should be validated with field tests
---
## Database Access
The athlete's training data is stored in SQLite at `~/.claude-coach/coach.db`. Query it using the built-in query command:
```bash
npx claude-coach query "YOUR_QUERY" --json
```
This works on any Node.js version (uses built-in SQLite on Node 22.5+, falls back to CLI otherwise).
**Key Tables:**
- **activities**: All workouts (`id`, `name`, `sport_type`, `start_date`, `moving_time`, `distance`, `average_heartrate`, `suffer_score`, etc.)
- **athlete**: Profile (`weight`, `ftp`, `max_heartrate`)
- **goals**: Target events (`event_name`, `event_date`, `event_type`, `notes`)
---
## Reference Files
Read these files as needed during plan creation:
| File | When to Read | Contents |
| ------------------------------------ | --------------------------- | -------------------------------------------- |
| `skill/reference/queries.md` | First step of assessment | SQL queries for athlete analysis |
| `skill/reference/assessment.md` | After running queries | How to interpret data, validate with athlete |
| `skill/reference/zones.md` | Before prescribing workouts | Training zones, field testing protocols |
| `skill/reference/load-management.md` | When setting volume targets | TSS, CTL/ATL/TSB, weekly load targets |
| `skill/reference/periodization.md` | When structuring phases 提问、分享技巧或报告无法使用。按“有帮助”数量排序。