Transition MCP
AI-powered coaching for runners, cyclists, swimmers, and triathletes, enabling personalized workouts, training plan adaptation, performance analytics, and AI coaching via Claude, MCP clients, or HTTP.
README
Transition MCP
AI-powered coaching for runners, cyclists, swimmers, and triathletes. Get personalized workouts, training plan adaptation, performance analytics, and AI coaching — all accessible from Claude, MCP clients, or any HTTP client.
Powered by Transition
Quick Start
Try it right now — no API key needed:
curl "https://api.transition.fun/api/v1/wod?sport=run&duration=45"
For personalized features (your training plan, performance data, AI coach), you need a Transition account + API key. See Getting an API Key below.
Three Ways to Use This
1. Claude Code / OpenClaw Skill (Recommended)
The simplest option. Claude reads the skill file and calls the API directly — no binary, no MCP protocol, just HTTP. Also available on OpenClaw.
# Clone into your skills directory
git clone https://github.com/nftechie/transition-mcp.git ~/.claude/skills/transition-mcp
Set your API key:
# Add to your shell profile (~/.zshrc, ~/.bashrc, etc.)
export TRANSITION_API_KEY="tr_live_xxxxxxxxxxxxxxxxxxxxx"
Now in Claude Code, you can say things like:
- "What's my workout today?"
- "Show me my fitness trend for the last month"
- "Ask my coach if I should do intervals or rest today"
- "Generate me a 30-minute swim workout"
Claude reads SKILL.md and knows how to call every endpoint.
2. MCP Server (Claude Desktop, Cursor, etc.)
For MCP-compatible clients. The server wraps the Transition API as MCP tools and resources.
Install:
cd mcp
go build -o transition-mcp .
Configure Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"transition-mcp": {
"command": "/absolute/path/to/transition-mcp",
"env": {
"TRANSITION_API_KEY": "tr_live_xxxxxxxxxxxxxxxxxxxxx"
}
}
}
}
Restart Claude Desktop. You'll see 6 tools and 3 resources available:
Tools:
| Tool | Description |
|---|---|
get_todays_workout |
Today's scheduled workout |
get_week_plan |
7-day training plan |
adapt_plan |
Trigger plan adaptation |
get_fitness_metrics |
CTL/ATL/TSB performance data |
chat_with_coach |
AI endurance coach |
generate_workout |
Free Workout of the Day (no auth) |
Resources:
| Resource | Description |
|---|---|
transition://athlete/profile |
Goals, experience, preferences |
transition://athlete/recent-activities |
Last 14 days of workouts |
transition://athlete/performance |
FTP, thresholds, PMC data |
3. Direct API
Use the API from scripts, automations, or any HTTP client. See SKILL.md for complete endpoint documentation with curl examples.
# Get this week's workouts
curl -H "X-API-Key: $TRANSITION_API_KEY" \
"https://api.transition.fun/api/v1/workouts?start=2026-02-09&end=2026-02-15"
# Ask the AI coach a question
curl -X POST -H "X-API-Key: $TRANSITION_API_KEY" \
-H "Content-Type: application/json" \
-d '{"message": "My left knee hurts after long runs. Should I adjust my plan?"}' \
"https://api.transition.fun/api/v1/coach/chat"
Getting an API Key
- Download Transition (iOS/Android)
- Create an account and complete onboarding
- Go to Settings > API Keys
- Tap Generate New Key — you'll see a key starting with
tr_live_ - Copy it immediately (it's only shown once)
Free tier includes 100 read requests/day and 3 AI requests/day. Plenty for personal use.
Rate Limits
| Tier | Read Endpoints | AI Endpoints |
|---|---|---|
| Free | 100/day | 3/day |
| Paid | 10,000/day | 100/day |
Read: workouts, metrics, profile, history | AI: coach chat, adapt, generate
Examples
See the examples/ directory for scripts:
curl-examples.sh— All endpoints with curl- More coming soon
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。