Email Processing MCP Server

Name: Email Processing MCP Server
Author: Cam10001110101

@Cam10001110101

代码仓库

关于 Email Processing MCP Server

暂无概览

#email-processing #mcp-server #outlook-integration

基本信息

分类

沟通协作

运行时

python

传输方式

stdio

发布者

Cam10001110101

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "mcp-server-outlook-email": {
      "command": "uv",
      "args": [
        "venv",
        ".venv"
      ]
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Email Processing MCP Server?

A cross-platform MCP server that processes Microsoft Outlook emails, generates vector embeddings using Ollama, and provides semantic search capabilities. It works on Windows, macOS, and any platform via Microsoft Graph API, and complies with the MCP 2025-06-18 specification.

How to use Email Processing MCP Server?

Install dependencies with uv, configure environment variables (MONGODB_URI, SQLITE_DB_PATH, etc.), set up provider-specific prerequisites (Outlook, Azure AD app), and add the server to Claude for Desktop’s claude_desktop_config.json. Tools like process_emails are invoked via natural language commands in Claude.

Key features of Email Processing MCP Server

Process emails from Outlook with date range filtering
Store emails in SQLite and embeddings in MongoDB
Semantic search across email content via vector embeddings
Multi-mailbox and multi-account support (Inbox, Sent, Deleted Items)
Cross-platform: Windows, macOS, Graph API (any platform)
12+ MCP tools: search, sentiment analysis, export, folder management

Use cases of Email Processing MCP Server

Search and analyze email archives using natural language queries
Automate email processing workflows with AI assistants (e.g., Claude)
Extract actionable items and contacts from Outlook mailboxes
Export email data to CSV, JSON, HTML, or Excel for reporting
Manage and organize emails across multiple accounts or tenants

FAQ from Email Processing MCP Server

What are the runtime dependencies?

Python 3.10+, Ollama with nomic-embed-text model, MongoDB server, and platform-specific components: Windows requires Outlook + pywin32, macOS requires Outlook for Mac, Graph API requires Azure AD app registration.

Where are email data and embeddings stored?

Raw emails and metadata are stored in a SQLite database (path set via SQLITE_DB_PATH). Vector embeddings are stored in MongoDB (connection string via MONGODB_URI).

How does the server determine which Outlook provider to use?

Set OUTLOOK_PROVIDER to auto (default) for automatic selection: Windows → COM automation, macOS → AppleScript, Linux/other → Graph API. Alternatively, force a provider with windows, mac, or graph.

Does the server support HTTP transport?

Yes. Run with --http flag to enable Streamable HTTP transport at http://localhost:8000/mcp, with full 2025-06-18 protocol compliance.

What authentication is required for Graph API access?

Register an Azure AD app with Mail.Read and User.Read.All application permissions, then provide GRAPH_CLIENT_ID, GRAPH_CLIENT_SECRET, and GRAPH_TENANT_ID as environment variables.