tractive-mcp
Enables AI assistants to query your pet's location, tracker status, position history, and more via the Tractive GPS pet tracker.
README
tractive-mcp
An MCP server for the Tractive GPS pet tracker. Lets AI assistants like Claude query your pet's location, tracker status, position history, and more.
Requirements
- Python 3.13+
- uv
- A Tractive account with at least one GPS tracker
Quick Start (from PyPI)
# Authenticate
uvx tractive-mcp auth
# Add to Claude Code
claude mcp add tractive -- uvx tractive-mcp
Install from Source
git clone https://github.com/rexchoppers/tractive-mcp.git
cd tractive-mcp
uv sync
uv run tractive-mcp auth
claude mcp add tractive -- uv run tractive-mcp
Authentication
Tractive does not provide any OAuth mechanisms. As a result, email and passwords have to be stored locally. Run the auth command to save your Tractive credentials:
uvx tractive-mcp auth
Credentials are stored at ~/.config/tractive-mcp/credentials.json with 0600 permissions.
Usage with Claude Code
claude mcp add tractive -- uvx tractive-mcp
Usage with Claude Desktop
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"tractive": {
"command": "uvx",
"args": ["tractive-mcp"]
}
}
}
Available Tools
Discovery
| Tool | Description |
|---|---|
list_pets |
Lists all pets on the account with their name, type, and device_id |
Location
| Tool | Description |
|---|---|
get_pet_location(device_id) |
Current GPS coordinates, speed, altitude, accuracy |
get_pet_distance_from_home(device_id) |
Distance in metres from the pet's home location |
get_recent_positions(device_id, hours, include_points) |
Position history with summary stats (distance, bounding box, point count). Set include_points=True for the full breadcrumb trail |
Tracker
| Tool | Description |
|---|---|
get_tracker_status(device_id) |
Battery level, charging state, connection state, firmware |
set_buzzer(device_id, active) |
Turn the tracker buzzer on/off |
set_led(device_id, active) |
Turn the tracker LED on/off |
set_live_tracking(device_id, active) |
Toggle live tracking mode (more frequent GPS, more battery) |
Pet Info
| Tool | Description |
|---|---|
get_pet_profile(device_id) |
Full pet profile — breed, weight, birthday, activity goals |
Emergency
| Tool | Description |
|---|---|
lost_pet(device_id, enable_live_tracking, enable_buzzer, enable_led) |
All-in-one emergency tool. Returns location, distance from home, tracker status, and recent activity. Optionally activates live tracking, buzzer, and LED |
Project Structure
src/tractive_mcp/
server.py — MCP server, tool definitions, auth CLI
client.py — Credential management, Tractive client helper
models.py — Response dataclasses
License
See LICENSE.
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。