Tableau MCP Server
Enables interaction with Tableau Cloud and Tableau Server through the REST API, supporting workbook management, view queries, extract refreshes, and content search operations.
README
Tableau MCP Server
A Model Context Protocol (MCP) server for Tableau Cloud and Tableau Server operations.
Project Status
Current Phase: Phase 6 Complete ✅ - Ready for Cloud Run Deployment
Completed Phases
- ✅ Phase 1: Project Setup
- ✅ Phase 2: Tableau API Client (13 methods)
- ✅ Phase 3: MCP Server Core (SSE transport, authentication)
- ✅ Phase 4: Core MCP Tools (6 tools)
- ✅ Phase 5: Advanced MCP Tools (3 tools)
- ✅ Phase 6: Cloud Run Deployment Infrastructure ✨ NEW!
Phase 6 Deliverables
- ✅ Optimized multi-stage Dockerfile
- ✅ Staging and production Cloud Run configurations
- ✅ Automated deployment scripts (
deploy-staging.sh,deploy-production.sh) - ✅ Secrets management automation (
create-secrets.sh) - ✅ Comprehensive deployment documentation (2,000+ lines)
- ✅ Automated testing script (
smoke-test.sh) - ✅ Monitoring and logging guides
Next Steps
- Deploy to staging: Run
./deploy-staging.sh(requires Google Cloud project) - Test deployment: Run
./smoke-test.sh staging - Phase 7: Create Cursor MCP configuration file
- Phase 8: Comprehensive testing and documentation
See PHASE_6_LOG.md for detailed Phase 6 implementation notes.
See PASS_OFF.md for complete project roadmap.
Project Structure
tableau-mcp-project/
├── src/
│ ├── server.ts # Main MCP server (Phase 3)
│ ├── tableau-client.ts # Tableau API client (Phase 2)
│ ├── tools/ # MCP tool implementations (Phase 4-5)
│ │ ├── list-workbooks.ts
│ │ ├── list-views.ts
│ │ ├── query-view.ts
│ │ ├── refresh-extract.ts
│ │ ├── search-content.ts
│ │ └── get-metadata.ts
│ └── types.ts # TypeScript type definitions
├── package.json
├── tsconfig.json
├── Dockerfile # Cloud Run deployment (Phase 6)
└── env.example # Environment variable template
Development Scripts
# Install dependencies
npm install
# Run in development mode with hot reload
npm run dev
# Build for production
npm run build
# Run production build
npm start
# Clean build artifacts
npm run clean
Quick Start - Cloud Run Deployment (Phase 6)
Prerequisites
- Google Cloud Project with billing enabled
- gcloud CLI installed and authenticated
- Docker installed locally
- Tableau Personal Access Token
Deploy to Staging
- Create secrets:
chmod +x create-secrets.sh
./create-secrets.sh staging
- Deploy service:
chmod +x deploy-staging.sh
./deploy-staging.sh
- Test deployment:
chmod +x smoke-test.sh
./smoke-test.sh staging
Deploy to Production (When Ready)
# Create production secrets
./create-secrets.sh production
# Deploy to production
chmod +x deploy-production.sh
./deploy-production.sh
# Test production deployment
./smoke-test.sh production
Comprehensive Guides
- Deployment: See
DEPLOYMENT_GUIDE.md - Secrets: See
SECRETS_SETUP.md - Testing: See
TEST_DEPLOYMENT.md - Monitoring: See
MONITORING_GUIDE.md - Phase 6 Log: See
PHASE_6_LOG.md
Environment Configuration
Copy env.example to .env and configure:
TABLEAU_SERVER_URL=https://your-server.tableau.com
TABLEAU_SITE_ID=your-site
TABLEAU_TOKEN_NAME=your-token-name
TABLEAU_TOKEN_VALUE=your-token-value
MCP_API_KEY=secure-random-key-here
PORT=8080
TABLEAU_API_VERSION=3.21
Dependencies
Core Dependencies
@modelcontextprotocol/sdk(^1.0.4) - MCP Server SDKexpress(^4.21.2) - HTTP server for SSE transportaxios(^1.7.9) - HTTP client for Tableau REST APIzod(^3.24.1) - Schema validationdotenv(^16.4.7) - Environment variable management
Development Dependencies
typescript(^5.7.2) - TypeScript compilertsx(^4.19.2) - TypeScript execution and hot reload@types/node&@types/express- Type definitions
Phase 1 Implementation Log
See PHASE_1_LOG.md for detailed implementation steps.
Next Implementation Phase
See PASS_OFF.md for Phase 2-8 implementation details.
License: MIT
Author: Agile Market Intelligence
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。