# Asian Recs [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/raymond-raymond/asianrecs-claude  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/asian-recs

## Description
SF and NYC restaurants, cafes, desserts and bars ranked by how Asian diners rate them.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "asian-recs": {
    "url": "https://asianrecs.com/api/mcp"
  }
}
```

## Documentation

## What the Asian Recs MCP server does

Asian Recs provides dining recommendations for San Francisco and New York. Its coverage includes restaurants, cafes, dessert spots, and bars. Results are ranked using ratings from Asian diners alongside ratings from the broader population, which supports questions about where Asian diners rate a place or dish more highly than everyone else.

The Asian Recs MCP server is suited to conversational discovery rather than general restaurant database administration. Example requests include finding soup dumplings in Flushing, locating boba near 24th and Mission, or identifying East Village ramen restaurants that Asian diners prefer relative to overall ratings.

## How it works

The service exposes an MCP endpoint at `https://asianrecs.com/api/mcp`. An MCP-compatible client can connect to that endpoint and let an AI assistant issue dining-related requests in natural language. The README specifically documents use with Claude.ai, Claude Desktop, and Claude Code.

The underlying data comes from asianrecs.com and is refreshed weekly. Recommendations should therefore be understood as being based on the service’s latest weekly data rather than live availability, current wait times, or real-time business status. The supplied material does not document reservation actions, ordering, navigation, or other operations beyond querying the recommendation data.

## Setup and configuration

Claude.ai and Claude Desktop users can find Asian Recs in the connectors directory. The README also gives a custom-connector path: open Settings, choose Connectors, select Add custom connector, and enter `https://asianrecs.com/api/mcp`.

Claude Code users can install the published plugin through the marketplace commands below:

```text
/plugin marketplace add raymond-raymond/asianrecs-claude
/plugin install asianrecs@asianrecs
```

No environment variables or additional configuration values are specified in the provided material. The documented connection uses the hosted MCP URL, so the setup does not describe a local database, API key, or self-hosted process.

The Asian Recs MCP server is a fit when the client can connect to a hosted MCP endpoint or install the documented Claude Code plugin. For other clients, confirm that they support the relevant MCP connection method before configuring the URL.

## Tools and capabilities

- Search for dining recommendations in San Francisco and New York.
- Explore restaurants, cafes, desserts, and bars.
- Compare Asian diner ratings with ratings from everyone else.
- Answer neighborhood- and landmark-based requests.
- Find recommendations for specific dishes or drink categories, such as ramen, soup dumplings, and boba.

The source material describes these capabilities through user-request examples rather than listing individual MCP tool names. It also does not specify filters, pagination, booking functions, or write operations.

_Full upstream README: https://allmcps.com/mcp/asian-recs/readme_