@striderlabs/mcp-lyft
MCP server connector for Lyft ride-sharing — request rides, get fare estimates, and track trips via browser automation.
README
@striderlabs/mcp-lyft
MCP server connector for Lyft ride-sharing — request rides, get fare estimates, and track trips via browser automation.
Installation
npm install -g @striderlabs/mcp-lyft
Or run directly with npx:
npx @striderlabs/mcp-lyft
MCP Configuration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"lyft": {
"command": "npx",
"args": ["@striderlabs/mcp-lyft"]
}
}
}
Tools
status
Check current session and route status.
Returns: Login state, user info, and saved pickup/destination.
login
Authenticate with your Lyft account.
| Parameter | Type | Required | Description |
|---|---|---|---|
identifier |
string | Yes | Email or phone number |
password |
string | Yes | Lyft account password |
headless |
boolean | No | Run browser headlessly (default: true) |
Note: If Lyft requires a verification code (OTP), you must complete sign-in manually in a browser first.
logout
Clear saved session, cookies, and route data.
set_pickup
Set the pickup location for your ride.
| Parameter | Type | Required | Description |
|---|---|---|---|
location |
string | Yes | Address or place name (e.g. "Times Square, NYC") |
set_destination
Set the destination for your ride.
| Parameter | Type | Required | Description |
|---|---|---|---|
location |
string | Yes | Address or place name (e.g. "JFK Airport") |
get_fare_estimate
Get fare estimates for the current pickup/destination route.
Requires: Both pickup and destination to be set via set_pickup and set_destination.
Returns: List of ride types with estimated fares and ETAs.
get_ride_options
Get available Lyft ride types for the current route.
Requires: Login + both pickup and destination set.
Returns: Available options (Lyft, Lyft XL, Lux, Lux Black, etc.) with prices and wait times.
request_ride
Request a Lyft ride. Returns a preview by default — set confirm=true to actually book.
| Parameter | Type | Required | Description |
|---|---|---|---|
ride_type |
string | No | Ride type (default: "Lyft") |
confirm |
boolean | No | Set true to book (default: false — preview) |
Requires: Login + both pickup and destination set.
get_ride_status
Get the status of your current or most recent Lyft ride.
Requires: Login.
Returns: Driver info, ETA, vehicle details, and trip status.
cancel_ride
Cancel a pending or active ride.
| Parameter | Type | Required | Description |
|---|---|---|---|
reason |
string | No | Optional cancellation reason |
Note: Cancellation fees may apply depending on timing.
get_ride_history
Get recent ride history.
| Parameter | Type | Required | Description |
|---|---|---|---|
limit |
number | No | Number of rides to return (default: 10) |
Returns: List of past rides with date, route, fare, and status.
Typical Workflow
1. login → Authenticate with Lyft
2. set_pickup → "123 Main St, San Francisco, CA"
3. set_destination → "SFO Airport"
4. get_fare_estimate → See prices for all ride types
5. request_ride → Preview the ride (confirm=false)
6. request_ride → confirm=true to actually book
7. get_ride_status → Track your driver
8. cancel_ride → Cancel if needed
Session Storage
Session data is stored locally at ~/.striderlabs/lyft/:
| File | Contents |
|---|---|
cookies.json |
Browser session cookies |
auth.json |
Login metadata (identifier, time) |
route.json |
Saved pickup/destination |
Technical Details
- Uses Playwright (Chromium) for browser automation
- Stealth techniques applied to avoid bot detection
- Cookies persist across MCP calls for seamless sessions
- Default geolocation: New York City
- User-agent: Chrome 120 on macOS
Important Notes
- OTP/Verification: Lyft frequently requires SMS or email verification codes. If prompted, complete sign-in manually in a browser — cookies will be reused.
- CAPTCHA: If a CAPTCHA appears, complete it manually in a browser to establish a session.
- Cancellation Fees: Lyft may charge a fee if you cancel after a driver has been assigned.
- Mobile vs Web: Some Lyft features may only be available in the mobile app and not accessible via browser automation.
License
MIT © Strider Labs
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。