原型 · 预览版:收录官方注册中心里的全部服务器,每 30 分钟同步一次。评价和账户功能将在后续阶段推出。
PopNx MCP为使用 AI 的人准备的 MCP 服务器 · 用大白话讲清楚

Sauce API MCP

作者:saucelabs版本 1.0.0i★ –i在你的电脑上运行
注册名称io.github.saucelabs-sample-test-frameworks/sauce-api-mcp

安装

概览

作者提供的说明(英文原文)

An open-source MCP server that provides LLM access to the Sauce Labs API

PopNx MCP 的三大支柱

这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。

它能帮你做什么

我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。

  • 连接方式: 在你的电脑上运行
  • 传输方式: stdio
  • 分类: 搜索与 AI 记忆

安装

选择你使用的 AI 应用,然后复制设置。

安装之前

  • 基础检查:1 项中通过 0 项
  • 在你的电脑上运行,拥有你的用户账户的权限(文件、网络、应用)。
  • 需要你自己填写: SAUCE_USERNAME, SAUCE_ACCESS_KEY

你需要

  • PyPI已安装 Python 的 uv(用于 uvx 命令)

你用的是哪个应用?

在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。

打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json

{
  "mcpServers": {
    "sauce-api": {
      "command": "uvx",
      "args": [
        "sauce-api-mcp"
      ],
      "env": {
        "SAUCE_USERNAME": "your-value",
        "SAUCE_ACCESS_KEY": "your-value"
      }
    }
  }
}

在终端中运行这条命令

claude mcp add sauce-api --env SAUCE_USERNAME=your-value --env SAUCE_ACCESS_KEY=your-value -- uvx sauce-api-mcp

运行前请先填写空着的值。

在 Cursor 中打开

或者自己复制设置

粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json

{
  "mcpServers": {
    "sauce-api": {
      "command": "uvx",
      "args": [
        "sauce-api-mcp"
      ],
      "env": {
        "SAUCE_USERNAME": "your-value",
        "SAUCE_ACCESS_KEY": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。

在 VS Code 中打开

或者自己复制设置

粘贴到项目中的 .vscode/mcp.json

{
  "servers": {
    "sauce-api": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "sauce-api-mcp"
      ],
      "env": {
        "SAUCE_USERNAME": "your-value",
        "SAUCE_ACCESS_KEY": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。

这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。

你的选择通过 Cookie 保存在这台设备上,无需账户。

需要你填写的值

  • SAUCE_USERNAME env
  • SAUCE_ACCESS_KEY env

这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。

它能做什么

它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。

基础检查i

我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日

  • ! 注意

    作者与仓库一致

    发布者(saucelabs-sample-test-frameworks)不是仓库所有者(saucelabs);可能是其他人代为上架的

  • … 尚未检查

    仍在维护

    我们还没有从 GitHub 读取这个仓库

  • … 尚未检查

    有许可证

    我们还没有从 GitHub 读取这个仓库

  • – 不适用

    端点有响应

    它在你的电脑上运行,没有互联网端点

  • – 不适用

    使用 OAuth 登录

    在你的电脑上运行,改用本地设置

  • … 尚未检查

    工具描述中没有隐藏指令

    它在你的电脑上运行;我们只能读取互联网端点的工具列表

与“搜索与 AI 记忆”中的其他服务器比较i

  • 星标数多于此分类中其他 7250 个服务器的 0%
  • 发布到注册中心的时间比此分类中其他 7250 个服务器的 0% 更新
  • 基础检查结果等于或优于此分类中其他 7250 个服务器的 33%

与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。

常见问题

Sauce API MCP 安全吗?

它在 1 项适用的基础检查中通过了 0 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。

它需要哪些密钥或代码?

注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。

它在我的电脑上运行还是在云端?

它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。

它还在维护吗?

GitHub 仓库最后更新于–,最新版本(1.0.0)于14个月前进入注册中心。

哪些应用可以使用它?

支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Python (uv)。

有什么可以替代它?

同一分类中有:Context7, Codebase Memory, Exa

替代选择

同一分类中其他作者的服务器,按星标数排序。

如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i

Context7

作者:upstash

Up-to-date docs for AI agents

电脑或云端★ 64,224MIT今天更新
安装

Codebase Memory

作者:DeusData

Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.

在你的电脑上运行★ 48,148MIT20天前更新
安装

Exa

作者:exa-labs

Connect AI agents to Exa for web search, content fetching, and multi-step research.

在云端运行★ 5,213MIT上个月更新
安装

作者的 README

打开 README

按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。

Sauce Labs MCP Server

A Model Context Protocol (MCP) server that provides comprehensive integration with the Sauce Labs testing platform. This package includes two complementary MCP servers enabling AI assistants to interact with Sauce Labs' device cloud, manage test jobs, analyze builds, and monitor testing infrastructure through natural language conversations.

[!TIP]

Prefer a hosted, more feature-rich option? Sauce MCP is the

official MCP server hosted by Sauce Labs at https://mcp.saucelabs.com. There is nothing to install or update, and it

offers a broader tool set than this package, including live real-device control (tap, type, screenshot, app

install/launch), network capture, and AI-powered test case authoring, execution, and scheduling. See

Connect your AI client for setup instructions.

Servers

This package provides two separate MCP servers optimized for different use cases:

sauce-api-mcp (Core Server) — Full Sauce Labs API integration for account management, device discovery, job analysis, builds, storage, and tunnels.

sauce-api-mcp-rdc (RDC Server) — Real Device Cloud (RDC) focused server. Dynamically generates MCP tools at startup from the official Sauce Labs OpenAPI spec using FastMCP with an OpenAPIProvider, so the tool set is always up-to-date without code changes. Includes a small set of handwritten tools for endpoints that need special handling (binary payloads, session lifecycle, app installation polling).

Both servers can be configured simultaneously in your LLM client for full Sauce Labs coverage.

Features

🚀 Core Capabilities
  • Account Management: View account details, team information, and user permissions
  • Device Cloud Access: Browse 300+ real devices (iOS, Android) and virtual machines
  • Test Job Management: Retrieve recent jobs, analyze test results, and debug failures
  • Build Monitoring: Track build status, view job collections, and analyze test suites
  • Storage Management: Manage uploaded apps and test artifacts
  • Tunnel Monitoring: Check Sauce Connect tunnel status and configuration
🔧 Advanced Features
  • Real-time Device Status: Monitor device availability and usage across data centres
  • Cross-platform Testing: Support for both Virtual Device Cloud (VDC) and Real Device Cloud (RDC)
  • Test Analytics: Detailed job information including logs, videos, and performance metrics
  • Team Collaboration: Multi-team support with proper access controls
  • Dynamic RDC API: sauce-api-mcp-rdc auto-discovers the latest RDC v2 endpoints from the OpenAPI spec at startup; the cached spec is used as a fallback when the network is unavailable
  • Response shaping: Large API list responses are automatically truncated to keep LLM context budget under control ( configurable via SAUCE_MCP_MAX_RESPONSE_ITEMS)
  • File safety: File push/pull operations on devices are restricted to ~/.sauce-mcp/files/ to prevent path traversal

Prerequisites

  • Python 3.10+
  • pip
  • Sauce Labs account with API access
  • Claude Desktop, Gemini CLI, Goose, or another MCP-compatible LLM client

Installation

Install the package from PyPI:

pip install sauce-api-mcp

This installs both servers and registers their command-line entry points:

  • sauce-api-mcp — core server
  • sauce-api-mcp-rdc — RDC OpenAPI server

Verify installation:

which sauce-api-mcp

which sauce-api-mcp-rdc

Configuration for LLM Clients

Claude Desktop (Mac / Linux / Windows)
  1. Locate your Claude Desktop config file:
  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
  • Linux: ~/.config/Claude/claude_desktop_config.json
  1. Find your Python installation's bin directory:
   python3 -c "import sys; print(sys.prefix + '/bin')"
  1. Add both servers to your config:
   {

"mcpServers": { "sauce-api-mcp-core": { "command": "/path/to/bin/sauce-api-mcp", "env": { "SAUCE_USERNAME": "your-sauce-username", "SAUCE_ACCESS_KEY": "your-sauce-access-key" } }, "sauce-api-mcp-rdc": { "command": "/path/to/bin/sauce-api-mcp-rdc", "env": { "SAUCE_USERNAME": "your-sauce-username", "SAUCE_ACCESS_KEY": "your-sauce-access-key" } } } }

  1. Restart Claude Desktop to load the servers.
Gemini CLI

Add to ~/.gemini/settings.json:

{

"mcpServers": { "sauce-api-mcp-core": { "command": "/path/to/bin/sauce-api-mcp", "env": { "SAUCE_USERNAME": "your-sauce-username", "SAUCE_ACCESS_KEY": "your-sauce-access-key" } }, "sauce-api-mcp-rdc": { "command": "/path/to/bin/sauce-api-mcp-rdc", "env": { "SAUCE_USERNAME": "your-sauce-username", "SAUCE_ACCESS_KEY": "your-sauce-access-key" } } } }

Goose

Add to ~/.config/goose/config.yaml:

sauce-api-mcp-core:

cmd: /path/to/bin/sauce-api-mcp description: Sauce Labs MCP (Core) enabled: true envs: SAUCE_USERNAME: your-sauce-username SAUCE_ACCESS_KEY: your-sauce-access-key type: stdio

sauce-api-mcp-rdc: cmd: /path/to/bin/sauce-api-mcp-rdc description: Sauce Labs MCP (RDC) enabled: true envs: SAUCE_USERNAME: your-sauce-username SAUCE_ACCESS_KEY: your-sauce-access-key type: stdio

Environment Variables (Alternative)

Instead of adding credentials to config files, you can export them as environment variables:

export SAUCE_USERNAME="your-sauce-username"

export SAUCE_ACCESS_KEY="your-sauce-access-key"

Then omit the env block from your config. Both servers will automatically pick them up.

Configuration

Required Environment Variables
VariableDescription
SAUCE_USERNAMEYour Sauce Labs username
SAUCE_ACCESS_KEYYour Sauce Labs access key (found in Account Settings)
Optional Environment Variables
VariableDefaultDescription
SAUCE_REGIONUS_WESTData centre region: US_WEST, US_EAST, EU_CENTRAL
SAUCE_MCP_MAX_RESPONSE_ITEMS100Maximum list items returned before truncation (RDC server)
Getting Your Sauce Labs Credentials
  1. Log into your Sauce Labs account
  2. Navigate to Account → User Settings
  3. Copy your Username and Access Key

Troubleshooting Installation

"Command not found: sauce-api-mcp"

The entry point script isn't in your PATH. Use the full path approach:

python3 -m pip list | grep sauce-api-mcp

python3 -c "import sys; print(sys.prefix + '/bin/sauce-api-mcp')"

"ENOENT: no such file or directory" in MCP client

The MCP client is using a different Python environment. Solutions:

  1. Use the full absolute path (recommended) — update your config with the path printed by python3 -c "import sys; print(sys.prefix + '/bin/sauce-api-mcp')", not just which sauce-api-mcp
  2. Module invocation (fallback):
   "command": "python3",

"args": ["-m", "sauce_api_mcp.main"]

Example Prompts

  • *"Show me my recent test failures"*
  • *"Find available iPhone 16 devices"*
  • *"Analyse the performance of my latest build"*
  • *"Open a session on a Samsung device and install my app"*
  • *"What tunnels do I have running right now?"*

Available Tools

Core Server (sauce-api-mcp)
Account & Organisation
ToolDescription
get_account_infoRetrieve current user account information
lookup_usersFind users in your organisation
get_userGet detailed user information
lookup_teamsFind teams in your organisation
get_teamGet team details
list_team_membersList all members of a specific team
lookup_service_accountsList service accounts
get_service_accountGet service account details
get_my_active_teamGet the active team for the authenticated user
Device Management
ToolDescription
get_devices_statusList all devices and their current status
get_specific_deviceGet detailed information about a specific device
get_private_devicesList private devices available to your account
Test Jobs
ToolDescription
get_recent_jobsRetrieve your most recent test jobs
get_job_detailsGet comprehensive details about a specific job
get_real_device_jobsList active jobs on real devices
get_specific_real_device_jobGet details about a specific real device job
get_specific_real_device_job_assetDownload job assets (logs, videos, screenshots)
Builds
ToolDescription
lookup_buildsSearch for builds with filters
get_buildGet detailed information about a specific build
get_build_for_jobGet the build associated with a job
lookup_jobs_in_buildList all jobs within a build
Storage
ToolDescription
get_storage_filesList uploaded application files
get_storage_groupsList app storage groups
get_storage_groups_settingsGet settings for a storage group
upload_file_to_storageUpload an app file to Sauce Storage
update_storage_group_settingsUpdate app group settings (resigning, instrumentation, etc.)
Tunnels
ToolDescription
get_tunnels_for_userList active Sauce Connect tunnels
get_tunnel_informationGet details about a specific tunnel
get_current_jobs_for_tunnelSee how many jobs are using a tunnel
get_tunnel_version_downloadsGet download URLs for Sauce Connect versions
Test Assets & Logs
ToolDescription

| get_test_assets | Retrieve test artifacts for a VDC job

只显示了开头部分;其余内容请在仓库中阅读

问 PopNx