postcardbot-mcp-server
Send real physical postcards worldwide via AI agents. Supports single and bulk send (up to 500 recipients), balance checking, delivery tracking, and volume pricing from $0.72/card.
README
@postcardbot/mcp-server
<a href="https://glama.ai/mcp/servers/PostcardBot/postcardbot-mcp-server"> <img width="380" height="200" src="https://glama.ai/mcp/servers/PostcardBot/postcardbot-mcp-server/badge" alt="PostcardBot MCP server" /> </a>
Send real postcards from code. MCP server for Postcard.bot — let AI agents send physical postcards worldwide.
Works with Claude Desktop, Claude Code, Cursor, Windsurf, and any MCP-compatible client.
Quick Start
1. Get an API key
Sign up at postcard.bot, go to your account page, and generate an API key.
2. Add balance
Add credits at postcard.bot/buy-credits or via the API. Prepaid balance — volume pricing from $0.72/postcard.
3. Configure your MCP client
Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"postcardbot": {
"command": "npx",
"args": ["-y", "@postcardbot/mcp-server"],
"env": {
"POSTCARDBOT_API_KEY": "pk_live_your_key_here"
}
}
}
}
Claude Code:
claude mcp add postcardbot -- npx -y @postcardbot/mcp-server
Then set the environment variable POSTCARDBOT_API_KEY in your shell.
Cursor / Windsurf — add the same JSON config to your MCP settings.
Tools
send_postcard
Send a physical postcard. Cards are printed and shipped within 24 hours. Delivery takes 5-10 business days.
Parameters:
to— Recipient address (name,address_line1,cityrequired;address_line2,state,zip,countryoptional)from— Sender/return address (same fields)message— Back-of-card message (max 350 characters)image_url— Front image URL (publicly accessible, min 1875x1275px recommended)
Example prompt:
"Send a postcard to Jane Doe at 123 Main St, San Francisco CA 94102 with a photo of the Golden Gate Bridge and the message 'Wish you were here!'"
bulk_send
Send the same postcard to multiple recipients at once. Same message, image, and return address for all cards.
Parameters:
recipients— Array of recipient addresses (max 500, same fields astoabove)from— Sender/return address (same for all postcards)message— Back-of-card message (max 350 characters)image_url— Front image URL (publicly accessible, min 1875x1275px recommended)
Example prompt:
"Send a holiday postcard to my 3 friends: Jane at 123 Main St SF CA, John at 456 Oak Ave NYC NY, and Bob at 789 Pine Rd Austin TX"
Total cost is deducted upfront. Failed cards are automatically refunded.
check_balance
Check account balance, lifetime top-up amount, and current volume pricing tier.
get_pricing
Get all volume pricing tiers.
check_status
Check delivery status of a previously sent postcard.
Parameters:
postcard_id— The ID returned fromsend_postcard
REST API
You can also use the HTTP API directly without MCP:
curl -X POST https://postcard.bot/api/v1/postcards \
-H "Authorization: Bearer pk_live_your_key_here" \
-H "Content-Type: application/json" \
-d '{
"to": {"name": "Jane Doe", "address_line1": "123 Main St", "city": "San Francisco", "state": "CA", "zip": "94102"},
"from": {"name": "John Smith", "address_line1": "456 Oak Ave", "city": "New York", "state": "NY", "zip": "10001"},
"message": "Wish you were here!",
"image_url": "https://example.com/photo.jpg"
}'
Full API docs and OpenAPI spec: postcard.bot/developers
Environment Variables
| Variable | Required | Description |
|---|---|---|
POSTCARDBOT_API_KEY |
Yes | Your API key from postcard.bot/account |
POSTCARDBOT_API_URL |
No | Override API base URL (default: https://postcard.bot) |
Pricing
Volume pricing (based on lifetime top-up amount):
| Tier | Lifetime top-up | USA | International |
|---|---|---|---|
| Starter | $0-$49 | $1.99 | $2.99 |
| Bronze | $50-$99 | $1.49 | $2.49 |
| Silver | $100-$499 | $1.29 | $2.29 |
| Gold | $500-$999 | $0.99 | $2.15 |
| Platinum | $1,000-$4,999 | $0.85 | $2.05 |
| Diamond | $5,000+ | $0.72 | $1.99 |
Links
License
MIT
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。