PosDO MCP Server
Discovers over 356,000 restaurants across 20 countries, providing search, reviews, time-based offers, and cultural context via the Model Context Protocol.
README
<div align="center"> <img src="logo.png" alt="POS.DO MCP Server" width="200" height="200"/>
PosDO MCP Server
Discovery of 356K+ restaurants across 20 countries via the Model Context Protocol.
POS.DO is a discovery platform with global coverage of restaurants — names, addresses, reviews, photos, opening hours, cuisines, neighbourhoods, cultural context, and real-time availability with time-based offers. This MCP server exposes that catalogue to AI agents over the Model Context Protocol.
What this server provides
- 356,780 restaurants in 20 countries (US, ES, IT, MX, FR, CO, DE, GB, DO, PR, AU, CA, PT, JP, BE, CH, HK, VE, NZ, PY)
- 7 languages of content (ES, EN, IT, FR, DE, PT, CA)
- 14.1M+ user reviews with author, rating, text, language, photos
- Cultural context — nearby monuments, museums, parks, and heritage sites from Wikidata, per restaurant
- Flow time-based offers — off-peak discounts at participating restaurants, with real seat availability
- Reservation booking with idempotent semantics (safe retries via
Idempotency-Keyheader)
This is a proxy repository. The MCP endpoint runs at https://pos.do/mcp and is operated by EZZYPUSH SL. No local installation of the server is required — agents just point at the URL.
Quick start
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"pos-do": {
"url": "https://pos.do/mcp"
}
}
}
Restart Claude Desktop. The 7 tools below become available.
Cursor / Cline / Continue
Most MCP clients accept a remote HTTP server entry of the same shape. For Cursor:
{
"mcp.servers": {
"pos-do": {
"url": "https://pos.do/mcp",
"transport": "http"
}
}
}
Generic MCP client (raw JSON-RPC over HTTP)
curl -sS -X POST https://pos.do/mcp \
-H 'Content-Type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
Tools
| Tool | Purpose | Docs |
|---|---|---|
search_restaurants |
Search restaurants in a city, with cuisine/price/feature filters | docs |
get_restaurant_details |
Full profile for a restaurant by slug | docs |
get_restaurant_reviews |
Paginated reviews with star and date filters | docs |
get_flow_offers |
Today's time-based discount slots for a restaurant | docs |
book_flow_reservation |
Reserve a seat at a Flow-discounted slot | docs |
find_restaurants_near_landmark |
Restaurants within walking distance of a famous landmark | docs |
get_cultural_context_around_restaurant |
Monuments, museums, parks near a restaurant (Wikidata) | docs |
Full tool inventory with inputSchema is served live at /.well-known/mcp.json.
Endpoint and transport
- Production URL:
https://pos.do/mcp - Protocol: MCP 2024-11-05
- Transport: HTTP POST (JSON-RPC 2.0), Streamable HTTP compatible
- Discovery manifests:
https://pos.do/.well-known/mcp.json— official MCP discovery (fulltools[]withinputSchema)https://pos.do/.well-known/webmcp.json— webmcp.link manifesthttps://pos.do/.well-known/mcp/server-card.json— Smithery-style server cardhttps://pos.do/mcp/manifest.json— alias
Authentication and rate limits
The server runs in anonymous tier by default. No API key required.
| Tier | Per minute | Per day |
|---|---|---|
| Anonymous (default) | 60 req/min per IP | 10,000 req/day per IP |
| Reservation attempts | 10/min per IP | 5/day per email |
Higher-rate API keys are available on request — open an issue.
For idempotent reservations, send an Idempotency-Key HTTP header (any UUID or stable client-side key). Replays within 24 h return the original reservation instead of creating a duplicate.
Coverage and data sources
See docs/data-coverage.md for per-country counts and source attributions.
Examples
Status
Stable. Backwards-compatible changes only at this point. Breaking changes will be announced via CHANGELOG.md and bumped to 2.x.
License
Apache License 2.0 — see LICENSE.
Contact
- Issues / feature requests: github.com/posdo-io/mcp-server/issues
- Operator: EZZYPUSH SL (Spain)
- Website: pos.do
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.