Postfleet MCP
Gives AI agents their own email address with inbound parsing, classification, extraction, and prompt injection screening, plus tools to manage mailboxes, send/receive emails, and handle draft approval workflows.
README
Postfleet MCP
Give an AI agent its own email address. Postfleet is email infrastructure for agents — inbound is parsed, classified, extracted to your schema, and screened for prompt injection before your agent reads a word.
This repo holds the server source and its plugin manifests. It is a mirror — the code is
developed in the Postfleet application repo and published to npm as
@postfleet/mcp; open issues here, send patches
here, and they land upstream. The server ships two ways:
| Hosted (no install) | https://api.postfleet.ai/api/mcp, streamable HTTP, Authorization: Bearer pf_... |
| Local (stdio) | @postfleet/mcp via npx |
Get a pf_ API key at postfleet.ai.
Tools
list_mailboxes, create_mailbox, list_domains, send_email, reply_email, list_inbox,
read_email, wait_for_email — plus the draft lifecycle (create_draft, list_drafts,
get_draft, update_draft, send_draft, delete_draft) for flows where a human approves
before mail goes out. A send that lands in that gate returns 202 pending_approval, which is
not a failure.
Every tool is thin-over-REST — it calls the same /api/v1/ endpoints your key already reaches
(OpenAPI), so auth, scoping, and quotas are enforced once.
Install
Cursor
The manifest in this repo declares both transports. To wire it up by hand, add your key:
{
"mcpServers": {
"postfleet": {
"url": "https://api.postfleet.ai/api/mcp",
"headers": { "Authorization": "Bearer pf_your_key_here" }
}
}
}
Claude Code
claude mcp add postfleet --env POSTFLEET_API_KEY=pf_your_key_here -- npx -y @postfleet/mcp
Any stdio client
{
"mcpServers": {
"postfleet": {
"command": "npx",
"args": ["-y", "@postfleet/mcp"],
"env": { "POSTFLEET_API_KEY": "pf_your_key_here" }
}
}
}
POSTFLEET_API_URL (default https://api.postfleet.ai) overrides the API origin for staging or
self-hosted deployments.
Keys
Keys are scoped — bootstrap can only create mailboxes, api covers mail operations, mcp is the
one to hand a client. Use a separate key per workload so you can revoke one without breaking the
rest. Never commit a key; the manifests in this repo deliberately contain none.
Links
- Registry listing:
ai.postfleet/postfleeton the MCP Registry - npm:
@postfleet/mcp - Docs: postfleet.ai/docs/mcp
Build from source
npm install && npm run build # tsup → dist/
MIT licensed.
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。