IM

inqvey/mcp

Developer tools

Эта страница переведена автоматически. Технические детали смотрите в оригинале. Открыть оригинал

0 stars 0 forks Качество 68 Тренд 0MIT

Inqvey MCP server - run real market research in Claude Desktop, Cursor, and other AI tools

Обзор

Run real quantitative market research directly in Claude Desktop, Cursor, Windsurf, ChatGPT Desktop, and other AI tools via the Model Context Protocol. Inqvey gives your AI assistant a real market research engine. Instead of guessing, it returns actual percentage distributions for any question you ask your target audience. Results are: - validated against Ipsos human panel studies - Equivalent to surveying ~1,000 respondents - Delivered in ~5 minutes per question (15-20 min for a typical 3-question research run) : demand validation, pricing research, purchase intent, feature prioritization, message testing, audience analysis. Your AI can guess at market research questions from general knowledge. But guesses are not data. Inqvey runs an actual simulation calibrated against real human survey panels, so you get statistical percentages you can use to make product, pricing, or marketing decisions. 1. Sign up at inqvey.com (free, no credit card) 2. Go to inqvey.

README

Показан фрагмент оригинального README. Открыть оригинал.

Inqvey MCP Server

Run real quantitative market research directly in Claude Desktop, Cursor, Windsurf, ChatGPT Desktop, and other AI tools via the Model Context Protocol.

What it does

Inqvey gives your AI assistant a real market research engine. Instead of guessing, it returns actual percentage distributions for any question you ask your target audience. Results are:

  • ±2-6pp accuracy validated against Ipsos human panel studies
  • Equivalent to surveying ~1,000 respondents
  • Delivered in ~5 minutes per question (15-20 min for a typical 3-question research run)

Use cases: demand validation, pricing research, purchase intent, feature prioritization, message testing, audience analysis.

Why MCP (vs asking your AI directly)

Your AI can guess at market research questions from general knowledge. But guesses are not data. Inqvey runs an actual simulation calibrated against real human survey panels, so you get statistical percentages you can use to make product, pricing, or marketing decisions.

Quick start

1. Get an API key

  1. Sign up at inqvey.com (free, no credit card)
  2. Go to inqvey.com/app/account/api-keys
  3. Click “Generate API Key”
  4. Copy the key - it’s shown only once

Free plan includes 9 research questions. Paid plans: Starter ($49/mo, 60 questions, 3 audiences) and Pro ($99/mo, 180 questions, 10 audiences).

2. Install in your AI tool

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "inqvey": {
      "url": "https://inqvey.com/api/mcp/v1",
      "headers": {
        "Authorization": "Bearer YOUR_INQVEY_API_KEY"
      }
    }
  }
}

Restart Claude Desktop.

Cursor / Windsurf

Add to your MCP config (usually ~/.cursor/mcp.json or equivalent):

{
  "mcpServers": {
    "inqvey": {
      "url": "https://inqvey.com/api/mcp/v1",
      "headers": {
        "Authorization": "Bearer YOUR_INQVEY_API_KEY"
      }
    }
  }
}

ChatGPT Desktop

  1. Open Settings > Connectors > Advanced > Developer mode (enable)
  2. Add a new connector with URL https://inqvey.com/api/mcp/v1
  3. Set header Authorization: Bearer YOUR_INQVEY_API_KEY

Smithery

npx smithery@latest add inqvey

Paste your API key when prompted. Smithery updates your AI tool config automatically.

3. Try it

In your AI tool, say something like:

“Would remote workers pay $20/mo for an AI assistant that sorts Slack messages by priority?”

The AI will invoke Inqvey tools:

  1. Check your existing Market Twins
  2. Create one if needed (asks you for audience details or a URL)
  3. Run research (~15 min)
  4. Return real percentages

Available tools

Tool What it does
list_market_twins List your saved simulated audiences
create_market_twin Create a new audience from a description
extract_audience_from_url Suggest audience fields from a company URL
Открыть полный README на GitHub

Установка

npx smithery@latest add inqvey
Новичок в MCP? Начните с руководства на mcpskill.org →

Конфигурация

{ "mcpServers": { "inqvey": { "url": "https://inqvey.com/api/mcp/v1", "headers": { "Authorization": "Bearer YOUR_INQVEY_API_KEY" } } } }