MCP App Store
by KeenEthics
Cafeteria icon

Cafeteria MCP Server for Claude and ChatGPT

By Cafeteria, Inc.

Overview

Cafeteria collects answers from young consumers (Gen Next) through short question sessions in the Cafeteria app. This connector lets brand and research teams with a Cafeteria account query the responses their organization has access to, straight from Claude. What you can do: - Find what's most mentioned: ranked brands, products, people, shows and songs for a topic, with mention and unique-respondent counts, filterable by age, gender and US state. - Analyze named brands or topics: mention counts, unique respondents, age/gender/state breakdowns and sample verbatim responses for each, side by side. - Run read-only ClickHouse SQL against documented Cafeteria tables for slices the other tools don't cover (results capped at 1,000 rows). - When the data doesn't cover a question, see an offer for a custom research table and, only if you ask for it, send that request to the Cafeteria team. Requirements and limits: you need a Cafeteria account. Sign-in uses OAuth, and every query is scoped server-side to the tables your organization is licensed for. The data tools are read-only; the only write action notifies the Cafeteria team about a custom table you asked for. Results depend on the data available to your organization and the filters you choose.

How to connect

  1. Claude: Open the Cafeteria listing in the Claude connectors directory and choose Connect. Claude listing Cafeteria exposes 5 tools here.
  2. ChatGPT: Open the Cafeteria listing in the ChatGPT app directory and choose Connect. ChatGPT listing Cafeteria exposes 3 tools here.

Tools

mention_count

ChatGPT
Analyze specific entities by name: pass a list of keyword groups (one per entity) and get back, for each, the mention count, unique users, demographic breakdown (age bands, gender, state), and a sample of verbatim responses — all in a single call. Use this whenever the user names specific brands, products, people, shows, etc. and wants sentiment, context, or side-by-side comparison ('what do teens say about Tower 28 and Saie?', 'compare mentions of Nike vs Adidas'). Do NOT use this for open-ended discovery ('what's popular?') — use top_entities for that. Multi-word names require all words to appear in the same response. Always batch every entity you care about into one call.

query_cafeteria_tables

ChatGPT
Query the Cafeteria data tables directly with read-only SQL — the escape hatch for slices the curated tools don't cover (custom multi-table joins, unusual aggregations, ad-hoc time series). For ranked discovery ('top X'), use top_entities. For analyzing specific named entities ('what do people say about Nike?'), use mention_count — it returns demographics + sample verbatims in one call and is faster than composing the equivalent SQL by hand. Results are capped at 1000 rows. Tables must use their database prefix (e.g. normalization.ai_normalized_responses) — full catalog in schema://tables.

top_entities

ChatGPT
Discover what's popular: returns a ranked list of the most-mentioned entities (brands, products, people, shows, songs, etc.) from Gen Z responses, with mention counts and unique-user counts per term. Use this when you don't yet know which specific items to study — e.g., 'top beauty brands among teens', 'most-mentioned songs in California'. When you already know the specific brand/product/person and want sentiment or verbatim quotes, use mention_count instead — it returns demographics and sample responses per keyword in one call. Empty results means filters matched nothing (try widening — drop category, broaden query); it is NOT a data-access error.

custom_table_offer

Claude

mention_count

Claude

query_cafeteria_tables

Claude

request_custom_table

Claude

top_entities

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/cafeteria"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/cafeteria)