vies-smoother
SLA'd EU VAT number validation for AI agents via VIES, with caching and circuit-breaking to handle flaky upstream, returning valid/invalid/unavailable responses.
README
vies-smoother
SLA'd EU VAT number validation for AI agents. Wraps the European Commission's VIES service (authoritative but flaky -- rate limits, per-country outages, inconsistent errors) with:
- A 24-hour cache
- A per-country circuit breaker (stops hammering a member state that's currently down)
- An honest three-state answer, never a guess:
valid-- the number is registered and activeinvalid-- the number does not exist or is inactiveunavailable_upstream-- VIES is down and there's no cache; you are not charged for this
Free tier: 100 lookups per key, no card required. $0.01/lookup after that ($5 / 500-credit minimum top-up).
This repo is the MCP client only (a thin wrapper over the hosted REST API
at https://vantane.com) -- it doesn't reimplement the cache/circuit
breaker logic, which lives server-side so metering is centralized
regardless of how you connect.
Quickstart
Claim a free key:
curl -s -X POST https://vantane.com/keys/claim
Option A: hosted, no install
Point any MCP client directly at the hosted endpoint:
{
"mcpServers": {
"vies-smoother": {
"url": "https://vantane.com/mcp",
"headers": { "X-API-Key": "sk_live_..." }
}
}
}
Option B: run this locally (stdio)
pip install -r requirements.txt
export RAILS_API_KEY=sk_live_...
python server.py
MCP client config:
{
"mcpServers": {
"vies-smoother": {
"command": "python",
"args": ["/path/to/server.py"],
"env": { "RAILS_API_KEY": "sk_live_..." }
}
}
}
Tools
validate_vat(country_code, vat_number)-- single lookup, 1 creditvalidate_vat_batch(items)-- up to 50 per call, 1 credit each
REST API (if you'd rather not use MCP)
POST /keys/claim-- claim a free keyPOST /vies/validate-- single lookupPOST /vies/validate/batch-- batch, max 50GET /vies/status-- live VIES availability, cache hit rate, 7-day SLA (free, no key)POST /topup-- buy more credits (requires an existing key)
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。