Run Customer Studies from Claude, ChatGPT, or Cursor
Step-by-step guide to running synthetic customer Studies from your AI agent. Covers Claude, Claude Code, ChatGPT, Cursor, and VS Code with the Minds MCP server.
This guide shows how to run synthetic customer Studies directly from Claude, ChatGPT, or Cursor without leaving the agent surface. Setup takes a few minutes per agent. Once connected, your agent can build Audiences, create Studies, ask questions, and return findings as part of any workflow.
The Minds MCP server advertises a curated set of tools covering the whole loop, from building a grounded Audience to planning a multi-question Study and exporting results. Everything below assumes you have a Minds account. Sign up at getminds.ai if you don't.
What You Can Do Once Connected
Your agent gains the ability to:
- Build a grounded Audience of Minds from a plain-language brief
- Ask one Audience a standalone question
- Create a Study that compares several Audiences
- Plan a multi-question Study, review it, and run it after explicit confirmation
- Read a summary of the findings across completed questions
- Export results as PDF, DOCX, PPTX, or Markdown reports, or as CSV, XLS, or SPSS raw data
The agent calls these tools in sequence. A typical sequence: an agent receives a brief, builds an Audience, asks it a question or plans a short questionnaire, and returns a synthesized summary. Simple questions usually come back within a few minutes.
Setup for Claude Desktop
For the canonical product docs, keep the Minds MCP setup guide open alongside this walkthrough.
Claude Desktop and claude.ai share the same custom connectors, and you sign in with your Minds account over OAuth.
Step 1. Open Customize → Connectors, click +, and choose Add custom connector.
Step 2. Enter the URL https://getminds.ai/mcp and click Add.
Step 3. Click Connect, sign in to Minds, and choose Allow.
Step 4. In a conversation, enable Minds from + → Connectors and test by asking Claude "list my Minds Studies"; it should call list_studies and return your existing Studies.
If your setup cannot complete OAuth, Claude Desktop can instead reach Minds with an API key through the mcp-remote bridge. Generate a key in your account settings under API Keys (keys start with minds_) and add this entry to claude_desktop_config.json:
{
"mcpServers": {
"minds": {
"command": "npx",
"args": [
"-y", "mcp-remote",
"https://getminds.ai/mcp",
"--header",
"Authorization: Bearer minds_YOUR_API_KEY"
]
}
}
}
Setup for Claude Code
Claude Code connects to the remote server directly over HTTP.
Step 1. Run claude mcp add --transport http minds https://getminds.ai/mcp in your terminal.
Step 2. Start Claude Code, run /mcp, select minds, and choose Authenticate.
Step 3. Sign in to Minds in the browser window that opens and choose Allow. An API key is optional; add --header "Authorization: Bearer minds_YOUR_API_KEY" to the first command to use one instead.
Step 4. Test by asking Claude Code to list your Audiences.
Setup for ChatGPT
ChatGPT adds Minds as an MCP app with OAuth sign-in.
Step 1. In ChatGPT on the web, open Plugins and choose Add → Create MCP App. If the option is missing, turn on developer mode or ask your workspace admin to allow custom MCP apps.
Step 2. Name it Minds, set the server URL to https://getminds.ai/mcp and authentication to OAuth, then click Create.
Step 3. Click Continue to Minds, sign in with your Minds account, and choose Allow.
Step 4. In a new conversation, type @Minds and ask your question. The ChatGPT guide also covers the plugin directory and the desktop app, where you add Minds under Plugins → Add → Add MCP server.
Setup for Cursor and VS Code
Cursor and VS Code with GitHub Copilot both connect to the remote server directly and sign in with OAuth, so no bridge or API key is needed.
Step 1. In Cursor, add Minds to ~/.cursor/mcp.json:
{
"mcpServers": {
"minds": { "url": "https://getminds.ai/mcp" }
}
}
Step 2. Authenticate when Cursor prompts, sign in to Minds, and choose Allow.
Step 3. The Minds tools appear in the agent's tool picker.
In VS Code, run MCP: Add Server from the Command Palette, choose HTTP, and enter https://getminds.ai/mcp. Sign in when prompted, then use the tools from Copilot Chat in agent mode.
Running Your First Study
Once connected, the simplest workflow is:
Step 1. Ask the agent: "Build an Audience of mid-market B2B SaaS marketing managers in Germany."
The agent calls create_audience_from_brief. Minds researches the population and builds the Audience; the agent can follow progress with get_audience_creation_progress.
Step 2. Ask: "Ask this Audience whether they would pay 200 euros per month for a tool that automates customer interviews."
The agent calls ask_audience, which creates a private Study for that Audience, and reads the answers with get_study_status. The agent summarizes the responses.
Step 3. Ask: "Export this Study as a Markdown report."
The agent calls export_study and checks the export with get_study_status. You get a clean Markdown file with the full Study.
For a questionnaire, the agent calls plan_study_questions once with every question, shows you the plan, and only runs it with run_study_questions after you confirm that exact revision. You can chain the steps in a single prompt: "Build an Audience of US fintech early adopters, ask them about onboarding friction, and export the result." The agent runs the tools in sequence.
Available Tools
The Minds MCP server groups its advertised tools by product area:
Audience tools. list_audiences for browsing existing Audiences, create_audience_from_brief for building a grounded Audience, get_audience_creation_progress for following creation, ask_audience for asking one Audience a question, and export_audience and duplicate_audience for sharing and reuse.
Study tools. list_studies for browsing previous Studies, create_study for setting up a Study over several Audiences, ask_study for one standalone question, get_study_status for results, get_study_summary for the findings, and export_study for PDF, DOCX, PPTX, Markdown, CSV, XLS, or SPSS output.
Planning tools. plan_study_questions and run_study_questions for reviewed multi-question Studies, list_research_methods for methods such as MaxDiff, and Study drafts and templates for reuse.
The agent doesn't need to know these names. It reads the tool descriptions from the MCP handshake and chooses based on the task. You phrase your request naturally and the agent picks the right tool.
Patterns That Work Well
A few patterns we see succeed across teams:
Brief-to-summary in one prompt. Compose the entire research brief, the question, and the desired output format in a single prompt. The agent runs the full sequence.
Iterative messaging tests. Run a Study once with a baseline message, then ask the agent to test variants. The agent reuses the Study and just calls ask_study with new wording.
Comparative Studies. Create one Study with two Audiences (for example early adopters and late majority) and ask the same question of both. Minds reports each Audience's answers side by side.
Daily checks. Schedule the agent to run a brief daily Study question on a specific topic. The agent posts the result to Slack via a connected Slack MCP.
Auth and Cost
API keys use Bearer auth. Keys are scoped to a single Minds account. Revoke and rotate from your account dashboard.
OAuth flows for ChatGPT, Claude, Claude Code, Cursor, and VS Code use the same backend. Tokens expire and refresh automatically.
Pricing follows the public Minds plans: Free, Premium at 59 EUR/month, Team at 99 EUR per seat per month with a 1-seat minimum, and Enterprise custom pricing. Each tool call counts against the same quota as a manual Study run on the web app. See pricing for details.
Troubleshooting
If tools list but never fire, the most common cause is a client bug rather than a server issue. The Open WebUI integration in particular has known regressions in some versions, and on the Native function-calling path it requires Bearer auth (the upstream model provider cannot complete the OAuth handshake on Minds' behalf). Switch the connector to API key auth, or disable streaming on the model.
If OAuth fails on first connection, the most common cause is a stale connector entry. Delete and re-add. The Minds OAuth handler issues fresh tokens on each registration.
If the agent calls a tool but the result looks empty, check that you have at least one Audience or Study already created in your Minds account. Some tools, like list_studies, return an empty list if you have none.
What's Next
Once your agent is connected, the next step is composing it with other MCP servers. The most common combination is Minds plus an analytics MCP (PostHog or GA4) and a notification MCP (Slack). This lets the agent run a Study, validate the finding against real product data, and report the result to a team channel automatically. The full list of MCP servers worth connecting in 2026 is in our companion post on the best MCP servers for marketing and research agents. For reference, the live Minds MCP tools list shows every tool exposed to connected agents.
If you're new to the category, the definition of agentic market research gives the one-page overview of why agents change the workflow at all.
That's the full agentic research loop. It runs without a human in the middle, takes minutes instead of weeks, and costs a fraction of a traditional study. The setup above is the only thing standing between most marketing and product teams and that workflow.
