MCP Dataverse
The most complete MCP server for Microsoft Dataverse.
README
MCP Dataverse
<div align="center">
<img src="assets/logo.webp" alt="MCP Dataverse" width="180" />
The most complete MCP server for Microsoft Dataverse.
67 tools · 4 resources · 10 guided workflows · Zero config auth
</div>
Why MCP Dataverse?
AI agents hallucinate schema, guess column names, and build broken OData queries. This server gives them real-time access to your Dataverse environment — schema, records, metadata, solutions — through the Model Context Protocol.
- No Azure AD app registration — device code flow, zero pre-configuration
- Works with any MCP client — VS Code, Claude, Cursor, Windsurf, Gemini, Codex CLI
- Atomic tools — each tool does one thing well; the AI picks the right one
- Structured outputs — every response returns
{summary, data, suggestions} - Guardrails — destructive operations require explicit confirmation
- Encrypted tokens — AES-256-GCM cached credentials, never logged
Install
npx mcp-dataverse install
The interactive wizard configures your environment, registers the server in VS Code, and authenticates your Microsoft account in under 2 minutes.
Requires Node.js 20+. For other clients (Claude, Cursor, Windsurf…) see Multi-Client Setup.
Authentication
No PAC CLI, no app registration, no client secret. Uses Microsoft's device code flow (MSAL):
- First tool call → a sign-in code appears in the MCP Output panel (
View → Output → MCP) - Open
https://microsoft.com/devicelogin→ enter the code → sign in with your work account - Done. Token is cached encrypted — all future starts are silent
Re-authenticate after ~90 days of inactivity: npx mcp-dataverse-auth
Capabilities
| Category | Count | Description |
|---|---|---|
| Metadata | 8 | Tables, schema, relationships, option sets, entity keys |
| Query | 3 | OData, FetchXML, paginated retrieval |
| CRUD | 6 | Get, create, update, delete, upsert, assign |
| Actions & Functions | 6 | Bound/unbound Dataverse actions and functions |
| Batch | 1 | Up to 1000 operations atomically |
| Solutions | 3 | List solutions, components, publish customizations |
| Search | 1 | Full-text Relevance Search |
| Users & Teams | 3 | Users, roles, teams |
| Files | 2 | Upload/download file and image columns |
| + more | … | Audit, trace logs, delta tracking, impersonation, annotations… |
| Assistance | 4 | Tool router, workflow guide |
HTTP Transport
Run as an HTTP server for multi-client use:
MCP_TRANSPORT=http MCP_HTTP_PORT=3000 MCP_HTTP_SECRET=mysecret node dist/server.js
Connect using VS Code / Copilot with:
{
"servers": {
"dataverse": {
"type": "http",
"url": "http://localhost:3000/mcp",
"headers": {
"Authorization": "Bearer mysecret"
}
}
}
}
Troubleshooting
| Symptom | Fix |
|---|---|
| No sign-in prompt | Open View → Output → MCP — the device code is displayed there |
No MSAL accounts found |
Run npx mcp-dataverse-auth then restart the server |
Authentication timed out |
Restart the MCP server — a fresh code is generated automatically |
| Server not appearing in Agent mode | Run npx mcp-dataverse install or npx mcp-dataverse doctor |
| HTTP errors | Run npx mcp-dataverse doctor to diagnose config and connectivity |
Performance Tip
MCP Dataverse is designed to be comprehensive, but most AI models work best with fewer tools in context. Deselect the tools you don't need in your client's tool picker (e.g. VS Code Chat panel) to keep the agent focused and responsive.
Roadmap
| Version | Feature | Status |
|---|---|---|
| v0.4 | HTTP transport + attribute management + schema consistency | ✅ Released |
| v0.5 | Enterprise auth (Client Credentials, Managed Identity) + MCP Prompts | 🔴 Planned |
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。