mcp-supertours
Tour and excursion search on supertours.ru — destinations across Russia and abroad.
README
supertours/mcp-server-public
Public manifest and install instructions for the supertours.ru MCP server — a Model Context Protocol server that lets AI agents (Claude, Cursor, ChatGPT, Cline, Continue and others) search tours and excursions on supertours.ru by destination, price, duration, rating and more.
MCP endpoint: https://mcp.supertours.ru/sse
Landing & docs: https://mcp.supertours.ru/ · tools reference
Transport: legacy HTTP+SSE (MCP spec 2025-06-18)
Authentication: none — anonymous, read-only
What's exposed
Seven anonymous read-only tools:
| Tool | Purpose |
|---|---|
search_tours |
Filter tours by destination (settlement / country), category, price range (currency-aware), duration (hours), start time of day, activity level, kids-allowed flag |
get_tour |
Full description, schedule, photos, tariffs by slug or UUID |
get_trending_tours |
Most-viewed tours of the recent period |
get_tour_reviews |
User reviews, rating, trip date |
list_tour_categories |
Resolve user phrases like "wine tasting" to category IDs |
list_regions |
Russian regions with timezone (use to scope a search to a federal subject) |
list_settlements |
Autocomplete cities/towns to settlement IDs (Russian and international) |
Full input schemas and call examples on the tools reference page.
Install
Cursor
One-click install: open https://mcp.supertours.ru/ and press Install in Cursor.
Manual: add to ~/.cursor/mcp.json:
{
"mcpServers": {
"supertours": {
"url": "https://mcp.supertours.ru/sse",
"type": "sse"
}
}
}
VS Code (Continue, Cline)
One-click: open https://mcp.supertours.ru/ and press Install in VS Code.
Claude Desktop
Claude Desktop talks to remote SSE servers through the mcp-remote proxy.
Add to ~/.claude/claude_desktop_config.json (macOS / Linux) or
%APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"supertours-tours": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.supertours.ru/sse"]
}
}
}
ChatGPT (custom GPTs)
In the GPT builder → Configure → Add MCP server, paste:
https://mcp.supertours.ru/sse
Quick sanity check
# Open SSE long-poll and read the first frame:
curl -N https://mcp.supertours.ru/sse
# expected:
# id:<uuid>
# event:endpoint
# data:/mcp/message?sessionId=<uuid>
# Full handshake + tools/list (returns 7 tools):
npx --yes @modelcontextprotocol/inspector --cli \
https://mcp.supertours.ru/sse --transport sse --method tools/list
Example prompts
Once installed, try these in your agent:
- "Find a tour in Cappadocia under 300 EUR."
- "Recommend three trending tours in Russia for the weekend."
- "Show reviews for tour
ekskursiya-po-istoricheskomu-tsentru." - "What categories of tours are available?"
The agent should call list_settlements / list_tour_categories to resolve
names to IDs and then search_tours with the right filters.
Status & limits
- Prices and availability are indicative — confirm at supertours.ru before booking.
- Anonymous read-only; booking and chat-with-organizer endpoints will be added behind OAuth 2.1 in a follow-up release.
- Rate limits at the proxy layer (60 search calls / minute per IP).
Links
- Web: https://supertours.ru
- MCP landing: https://mcp.supertours.ru/
- Tools reference: https://mcp.supertours.ru/tools
- OAuth discovery (RFC 9728): https://mcp.supertours.ru/.well-known/oauth-protected-resource
- Privacy policy: https://supertours.ru/ru/documents/policy.pdf
- Contact: inbox@supertours.ru
License
MIT — see LICENSE. The server source itself is private
(part of taktilidu-server); this repository hosts the public install
manifest and configuration only.
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。