# Juno AI Customer Research

> Juno AI Customer Research is an MCP server for analytics and data. Run AI-moderated customer interviews and export transcripts from your coding agent. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Platforms:** API
- **Website:** https://www.heyjuno.co/documents/mcp/quickstart
- **Repository:** https://github.com/hey-juno/juno-agent-toolkit
- **MCP Registry name:** io.github.hey-juno/juno
- **Categories:** [analytics and data](https://ai-powered.com/categories/analytics-and-data)
- **Listing updated:** 2026-11-15

## About

Juno runs qualitative interviews with real people, and its toolkit's stance is that you reach for it when analytics, support tickets or past research leave a "why" unanswered. Through this hosted endpoint an agent can brief a study, review how the interview is set up, collect responses and pull evidence back out.

A first call such as create_study takes a plain-language guidance string, for example interviewing churned customers about why they left, and returns a job id and thread id. New studies start in Test mode, and the toolkit's workflow keeps drafting, going live, inspecting interviews and exporting as separate steps; it does not recruit or contact participants on its own.

Interview metadata does not include transcripts. Completed Live-mode interviews are retrieved as CSV exports under the organisation's permissions and redaction settings, and Test-mode runs are rehearsals rather than customer evidence. Connect with OAuth 2.1 (dynamic client registration, PKCE) or a Juno API key.

## Features

- **Study authoring:** Create a study from a short guidance prompt and get a job and thread id back.
- **Test mode first:** Studies begin in Test mode so the brief can be reviewed before going live.
- **Participant links:** The registry entry lists participant links as part of the toolset.
- **CSV exports:** Transcripts and answers for completed Live interviews come out as CSV exports.
- **OAuth or API key:** Connect with OAuth 2.1 plus PKCE, or a jsk_live / jsk_test bearer key.
- **Python client:** A small Python 3.11 client and example workflow ship in the toolkit repository.

## Good for

- Interview churned customers about why they left
- Probe onboarding drop-off with follow-up questions
- Collect first-hand explanations to sit beside product analytics
- Export interview answers to analyse in your own tools

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http juno https://api.heyjuno.co/v1/mcp/
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://api.heyjuno.co/v1/mcp/
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "juno": {
      "url": "https://api.heyjuno.co/v1/mcp/"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "juno": {
      "type": "http",
      "url": "https://api.heyjuno.co/v1/mcp/"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "juno": {
      "serverUrl": "https://api.heyjuno.co/v1/mcp/"
    }
  }
}
```

### Grok Bot (Grok Bot desktop app → Connect apps → Add a custom connector)

```sh
https://api.heyjuno.co/v1/mcp/
```

Enter a Name (for example "Juno AI Customer Research"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "juno": {
      "type": "http",
      "url": "https://api.heyjuno.co/v1/mcp/"
    }
  }
}
```

## Tools

- `create_study`: Create a study from a guidance prompt; returns a job id and thread id

## FAQ

### How do I connect?

Run claude mcp add --transport http juno https://api.heyjuno.co/v1/mcp/ for OAuth, or send a Juno API key as a bearer header.

### Do the interview records include transcripts?

No, interview metadata has none; use the CSV export workflow for completed Live-mode interviews.

### Does it contact participants automatically?

The README says the example workflow does not recruit or contact participants automatically.

### What is Test mode?

New studies start there; simulations are rehearsal material, not customer evidence.

### Is the Python toolkit on PyPI?

Not yet; the README says the preview is distributed from GitHub.

### How do I add Juno AI Customer Research to Claude Code?

Run this command in your terminal: claude mcp add --transport http juno https://api.heyjuno.co/v1/mcp/

### What are the best alternatives to Juno AI Customer Research?

The closest alternatives to Juno AI Customer Research are Usercall, Deutero, User Intuition, Askli Customer Research and GatherHQ.

## Where this comes from

- Description: written from the project's README and the project's website.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [Usercall](https://ai-powered.com/mcp/usercall): Usercall is a paid MCP server for analytics and data.
- [Deutero](https://ai-powered.com/mcp/deutero): Deutero is a freemium MCP server for analytics and data.
- [User Intuition](https://ai-powered.com/mcp/user-intuition): User Intuition is a paid MCP server for analytics and data.
- [Askli Customer Research](https://ai-powered.com/mcp/askli-customer-research): Askli Customer Research is a paid MCP server for analytics and data.
- [GatherHQ](https://ai-powered.com/mcp/gatherhq): GatherHQ is a freemium MCP server for analytics and data.
- [Versive](https://ai-powered.com/mcp/versive): Versive is a freemium MCP server for analytics and data.
- [SurveysAI Qualitative Analysis – interview and focus group analysis](https://ai-powered.com/mcp/surveysai-qualitative-analysis-interview-and-focus-group-ana): SurveysAI Qualitative Analysis – interview and focus group analysis is a paid MCP server for analytics and data.
- [Patchwell](https://ai-powered.com/mcp/patchwell): Patchwell is a paid MCP server for files and documents.

---
Source: https://ai-powered.com/mcp/juno-ai-customer-research
