mcp-wag

mcp-wag

An MCP server that enables AI assistants to interact with Wag pet care services using Playwright browser automation. It allows users to search for walkers, book one-time or recurring sessions, track walks via GPS, and retrieve pet report cards.

Category
访问服务器

README

@striderlabs/mcp-wag

An MCP (Model Context Protocol) connector for Wag dog walking and pet care services. Enables AI assistants like Claude to search for walkers, book walks, check walk status with GPS tracking, view walker profiles, leave reviews, and retrieve pet report cards — all via Playwright browser automation against the live Wag website.

What It Does

This MCP server wraps the Wag web app using Playwright so any MCP-compatible AI client can:

  • Search for available dog walkers, sitters, boarding, or daycare near a location
  • Book one-time or recurring walk sessions
  • Track an in-progress walk with GPS coordinates
  • View full walker profiles (bio, certifications, reviews, pricing, photos)
  • Rate and tip a walker after a completed walk
  • Retrieve the post-walk report card (photos, potty breaks, behavior notes, route)

Installation

npm install -g @striderlabs/mcp-wag
# Install Playwright browser binaries (one-time setup)
npx playwright install chromium

Or use it directly from the tarball:

npm install -g ./striderlabs-mcp-wag-1.0.0.tgz
npx playwright install chromium

Configuration

Set these environment variables before running the server:

Variable Required Description
WAG_EMAIL For auth tools Your Wag account email address
WAG_PASSWORD For auth tools Your Wag account password
WAG_HEADLESS No Set to false to show the browser window (for debugging)

search_walkers and view_walker_profile do not require credentials. All booking, status, rating, and report tools require WAG_EMAIL and WAG_PASSWORD.

Usage with Claude Desktop

Add the following to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "wag": {
      "command": "mcp-wag",
      "env": {
        "WAG_EMAIL": "your@email.com",
        "WAG_PASSWORD": "yourpassword",
        "WAG_HEADLESS": "true"
      }
    }
  }
}

If you installed from a local path instead of globally:

{
  "mcpServers": {
    "wag": {
      "command": "node",
      "args": ["/path/to/dist/index.js"],
      "env": {
        "WAG_EMAIL": "your@email.com",
        "WAG_PASSWORD": "yourpassword"
      }
    }
  }
}

Tools

search_walkers

Search for available dog walkers by location, date, and service type.

Input:

Field Type Required Description
location string Yes City, zip, or full address
date string Yes Date in YYYY-MM-DD format
service_type "walk" | "sitting" | "boarding" | "daycare" Yes Type of service
duration_minutes number No Desired walk duration (20, 30, 60)

Output: List of walker cards with name, rating, review count, price, availability, and profile URL.


book_walk

Book a dog walking session. Requires authentication.

Input:

Field Type Required Description
walker_id string Yes Walker profile slug or ID
pet_name string Yes Name of your pet
date string Yes Date in YYYY-MM-DD format
time string Yes Start time, e.g. "10:00 AM"
duration_minutes number Yes Duration in minutes
service_type string Yes Service type
special_instructions string No Notes for the walker

Output: Booking confirmation with booking ID, confirmation URL, and status message.


schedule_recurring

Set up a recurring walk schedule. Requires authentication.

Input:

Field Type Required Description
walker_id string Yes Walker profile slug or ID
pet_name string Yes Name of your pet
days_of_week string[] Yes Array of days, e.g. ["monday", "wednesday", "friday"]
time string Yes Walk time, e.g. "9:00 AM"
duration_minutes number Yes Duration in minutes
service_type string Yes Service type

Output: Confirmation of recurring schedule submission.


get_walk_status

Check the status of a booking and get GPS tracking if in progress. Requires authentication.

Input:

Field Type Required Description
booking_id string Yes Your booking ID

Output: Status, walker name, pet name, start/end times, distance walked, live GPS coordinates (if available).


view_walker_profile

View a walker's full public profile. No authentication required.

Input:

Field Type Required Description
walker_id string Yes Walker profile slug or ID

Output: Full profile including bio, location, rating, review count, pricing per service, certifications, services offered, photos, member since date, repeat client rate, response rate, and up to 5 recent reviews.


rate_walker

Submit a star rating and written review after a completed walk. Optionally add a tip. Requires authentication.

Input:

Field Type Required Description
booking_id string Yes Booking ID of the completed walk
rating number Yes Star rating 1–5
review string Yes Written review text
tip_amount number No Tip in dollars, e.g. 5.00

Output: Confirmation that the review was submitted.


get_pet_report

Retrieve the post-walk report card. Requires authentication.

Input:

Field Type Required Description
booking_id string Yes Your booking ID

Output: Full report card with walker name, pet name, walk date, duration, distance, steps, average pace, calories burned, potty breaks logged, behavior tags, walker notes, whether a route map is available, and walk photos.


Example Prompts

Once connected to Claude Desktop, you can use natural language:

  • "Find dog walkers near 90210 for next Saturday for a 30-minute walk"
  • "Show me the profile and reviews for walker john-smith-123"
  • "Book a walk with walker sarah-jones-456 for my dog Max on 2025-04-15 at 9 AM for 30 minutes"
  • "Set up recurring Monday/Wednesday/Friday walks at 8 AM with walker tom-brown-789"
  • "What's the status of my booking ABC123? Is the walker close?"
  • "Leave a 5-star review for booking XYZ789 saying 'Luna had a great time!' and add a $5 tip"
  • "Get the report card for booking DEF456 — did Bella have any potty breaks?"

Building from Source

git clone https://github.com/markswendsen-code/mcp-wag.git
cd mcp-wag
npm install
npm run build
# Output: dist/index.js

Troubleshooting

"WAG_EMAIL and WAG_PASSWORD environment variables are required" Set these in your shell or in the MCP server config's env block.

Login fails / redirected to login page

  • Verify your credentials by logging in to wagwalking.com manually.
  • Set WAG_HEADLESS=false to watch the browser and diagnose the issue.
  • Wag may require completing a CAPTCHA on first login from a new browser.

No walkers found in search results

  • The Wag website structure may have changed. Try setting WAG_HEADLESS=false to inspect.
  • Try a more specific location (full city + state, or zip code).
  • Ensure Playwright Chromium is installed: npx playwright install chromium.

Build fails

npm install
node build.js

Make sure Node.js 18+ is installed.

Playwright browser not found

npx playwright install chromium

License

MIT — Strider Labs

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选