email-mcp-postale

email-mcp-postale

Enables Claude Desktop to interact with Postale.io email accounts, supporting multiple mailboxes with features like reading, searching, replying, drafts, and batch operations.

Category
访问服务器

README

Email MCP Server for Postale.io

An MCP server that gives Claude Desktop / Cowork access to your Postale.io email mailboxes. Supports multiple accounts, reading, searching, replying, drafts, deleting, and batch operations.

Features

Tool Description
list_accounts Show all configured email accounts
list_folders Show folder structure of an account
get_unread_emails Fetch unread emails
get_recent_emails Fetch recent emails from the last X days
read_email Read a single email in full
search_emails Search emails (subject, sender, body)
send_email Send an email (with reply threading)
save_draft Save a draft (visible in Spark)
move_email Move an email to another folder
delete_email Move an email to trash
batch_delete_emails Delete multiple emails at once
mark_as_read Mark as read
mark_as_unread Mark as unread
get_mailbox_stats Stats: unread per folder
get_all_unread_summary Across ALL accounts: summarize unread

Setup

1. Clone and install dependencies

git clone https://github.com/ssk-machines/email-mcp-postale.git
cd email-mcp-postale
pip install -r requirements.txt
# or with uv:
uv pip install -r requirements.txt

2. Create configuration

cp config.json.example config.json

Then edit config.json — add an account entry for each mailbox:

{
  "accounts": [
    {
      "name": "My Account",
      "email": "me@example.com",
      "imap_host": "mail.postale.io",
      "imap_port": 993,
      "smtp_host": "mail.postale.io",
      "smtp_port": 465,
      "username": "me@example.com",
      "password": "your-password"
    },
    {
      "name": "Support",
      "email": "support@example.com",
      ...
    }
  ]
}

Tip: On Postale.io, the username is always the full email address.

3. Test

python server.py

If no errors appear, the server is running correctly in stdio mode.

4. Add to Claude Desktop

Open ~/Library/Application Support/Claude/claude_desktop_config.json and add:

With uv:

{
  "mcpServers": {
    "email": {
      "command": "uv",
      "args": [
        "--directory", "/path/to/email-mcp-postale",
        "run", "server.py"
      ]
    }
  }
}

Or with Python directly:

{
  "mcpServers": {
    "email": {
      "command": "python3",
      "args": ["/path/to/email-mcp-postale/server.py"],
      "env": {
        "EMAIL_MCP_CONFIG": "/path/to/email-mcp-postale/config.json"
      }
    }
  }
}

Replace /path/to/email-mcp-postale with the actual path where you cloned the repo.

5. Restart Claude Desktop

After saving the config, fully quit and reopen Claude Desktop. The email connector should appear as active under "Connectors".

Scheduled Tasks (Cowork)

Morning Briefing (daily, 7:00 AM)

Every morning at 7:00 AM:
1. Call get_all_unread_summary to fetch all unread emails across all accounts.
2. Categorize each email as:
   - 🔴 URGENT: Needs immediate reply (customers, partners, investors)
   - 🟡 REPLY NEEDED: Needs a reply, but not time-critical
   - 🟢 FYI: Just read, no action needed
   - ⚪ UNIMPORTANT: Newsletters, marketing, spam
3. Draft a reply for each 🔴 and 🟡 email using save_draft.
4. Mark ⚪ emails as read with mark_as_read.
5. Create a clear summary as a briefing.

Cleanup Task (weekly, Friday 5:00 PM)

Every Friday at 5:00 PM:
1. Search all accounts for newsletter and marketing emails from the past week.
2. Create a list for batch deletion.
3. Show me the list for confirmation before deleting.

Security Notes

  • Passwords are stored in plaintext in config.json — protect the file with chmod 600 config.json
  • Alternatively: use Postale.io app passwords (if supported)
  • Do NOT commit config.json to git

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选