iMario MCP Server已验证已精选

Ask anyone anything before you decide

iMario turns your own customer data and 5.4 billion people in 59 markets into synthetic audiences calibrated on real data. This server lets your AI assistant question them for you: put copy, a landing page, a price, a concept or an image in front of the audience it is for, and get back counted answers, themes and quotes in minutes.

Every audience is made of Synthetic Individuals: simulated people built from real human data, which a team or an AI agent can consult before a decision ships.

What you can ask

• "Ask US parents of toddlers which headline makes them more likely to click, and why: 'Sleep more, worry less' or 'The monitor that learns your baby'."
• "Run a page test on our pricing page with 30 US adults and tell me what would stop them signing up."
• "How would German adults aged 25 to 40 react to a $12 monthly price for this meal kit? Price the study first."
• "Ask US adults under 35 and over 55 whether they would trust an AI travel agent to book for them, and show me where they differ."

How a study runs

Step	Tool	Cost
Find an audience	list_audiences, get_audience	Free
Define a new one in a sentence	create_audience (previews before saving)	Free
Price the study	estimate_study	Free
Run it	run_study	Credits
Read the results	get_study	Free
Send your own image for an image test	request_upload, request_manual_upload	Free

upload_asset is used by the iMario upload card inside the chat; the assistant does not call it directly.

Research methods: a single question, a survey, interviews with follow-up probes, a focus group, a preference (A/B) test of text or images, a creative test on an ad or design, and feedback on a live web page.

What comes back

Counts, not predictions: how many people chose each option, the themes behind open answers with sample quotes, and for every answer whether it came from that person's own data or from a modelled profile. On iMario's public benchmark, synthetic audiences match real survey answer distributions with 89% accuracy, about 95% of what a rerun with real people achieves. Each study also opens in iMario, where you can see who said what and build a report.

Getting started

1. Add https://mcp.imario.ai/mcp to your MCP client.
2. Sign in to iMario, choose a workspace and approve the connection. Claude, ChatGPT, Claude Code and Codex do this in the browser; scripts and CI use a workspace API key instead.
3. Ask your first question.

In Claude Code, the companion plugin adds a skill that teaches the assistant to price a study and ask you before it spends:

/plugin marketplace add laugherxiao/imario-skills
/plugin install imario@imario

Good to know

• Listing audiences, pricing studies and reading results are free. Running a study costs one credit per answer, and new accounts start with 1,000 free credits.
• Workspace owners set a daily limit on what connected apps may spend.
• A viewer can read results but cannot run studies.

Documentation · Pricing · Plugin and skill