# Ask real people: human feedback, user testing, surveys

> Ask real people: human feedback, user testing, surveys is a freemium MCP server for agent tooling. MCP server that lets AI agents ask real people for reactions, approvals and answers. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Platforms:** API, CLI, Web
- **Website:** https://person.tools
- **Repository:** https://github.com/jonwashburn/person-tools
- **MCP Registry name:** tools.person/human-feedback
- **Categories:** [agent tooling](https://ai-powered.com/categories/agent-tooling)
- **Listing updated:** 2026-11-18

## About

The README describes a service where an AI agent sends a short task to several people, such as listening to audio, viewing a page, or choosing between two versions, and gets their answers back one at a time. Each answer comes with a record of what the person's page observed, such as how much audio played in real time and whether the page stayed in front. The README is explicit that this record does not prove the person actually listened or watched.

The service works through one MCP tool call that first returns a price quote, then opens the call under a spending cap the user sets. Simulated people are available free with a test key and no account, so developers can try the flow first. Real people are reached only through a panel they joined themselves via a link on their phone; strangers are not offered yet during the preview.

It suits developers and product teams whose agents cannot judge something like a song, a game build or a headline by themselves and need human reactions for testing or feedback. The site describes a separate consumer app for approvals on your own phone, but this listing covers the MCP server for agents asking people.

## Features

- **Price before spending:** Each call is quoted first, and nothing opens unless the total fits within the caller's max_total_usd and key caps.
- **Simulated test pool:** A free test key reaches simulated people, with every simulated answer marked as such.
- **Per-person activity record:** Each answer includes a record of playback, page focus, pauses and replays, and whether text was typed or pasted.
- **Own-panel invitations:** Developers can invite their own testers or customers to a panel joined from a link on their phones.
- **Open standard and suite:** The Ahoy Protocol is published as an Apache 2.0 draft, with a conformance suite that checks other servers against it.

## Good for

- Get a few people's reactions to a headline or product copy before shipping
- Have real listeners judge a generated audio clip
- Collect user testing feedback on a build or prototype
- Ask your own customers or testers a quick choice question from an agent

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http human-feedback https://api.person.tools/mcp
```

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

```sh
https://api.person.tools/mcp
```

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

```json
{
  "mcpServers": {
    "human-feedback": {
      "url": "https://api.person.tools/mcp"
    }
  }
}
```

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

```json
{
  "servers": {
    "human-feedback": {
      "type": "http",
      "url": "https://api.person.tools/mcp"
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "human-feedback": {
      "serverUrl": "https://api.person.tools/mcp"
    }
  }
}
```

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

```sh
https://api.person.tools/mcp
```

Enter a Name (for example "Ask real people: human feedback, user testing, surveys"), 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": {
    "human-feedback": {
      "type": "http",
      "url": "https://api.person.tools/mcp"
    }
  }
}
```

## Tools

2 of 6 tools say they're read-only. 4 of 6 tools say they can delete or change data. From the server's own tool annotations; not verified by us.

- `ask_people_quote` (says read-only): Prices a call without opening it: nobody is rung and nothing is spent. Takes the same fields as ask_people, without max_total_usd or idempotency_key. Nothing else is needed first: leave content out until your user gives you the material (the quote is priced from minutes), and leave pay out unless your user named one (left out, it is 0.00 on the simulated pool and on a panel; any pay you set must m
- `ask_people` (says it can change data): Opens a call: one short task to several people. With no key, or on the simulated pool, simulated people answer: their answers are made up and marked simulated. Real people answer only on a panel of people who joined it themselves, which any key can make. Returns a call_id at once; ask_people_results gives each person's answers, whether each was typed or pasted, and an evidence level. Use it only w
- `ask_people_results` (says read-only): Returns a call's answers so far, given the call_id from ask_people: first a summary (the state, a tally of each answer field, how many responses reached each evidence level), then one entry per person with their answers and whether each was typed or pasted, then the money (held_usd, spent_usd, returned_usd, and test_money, true throughout the preview: nobody receives real pay). detail: "full" adds
- `ask_people_cancel` (says it can change data): Stops a call. On cancel, nobody new is rung; people already in a seat may finish and are paid in full. Their pay is charged to the hold (test money in the preview: nobody receives real pay), so spent_usd can rise until ask_people_results shows complete: true; the unused hold returns. Returns the call's state and the money so far. Repeating it changes nothing; a cancel cannot be undone. Use it when
- `ask_people_send_outcome` (says it can change data): Sends the people who answered a call one line (at most 280 characters, and up to three lines a call) saying what you did with their answers, for example "Cut the intro from 40 to 12 seconds; most of you drifted there." The line is reviewed the way a call is and may be refused; if allowed, it appears on each answering person's receipt and they are notified. Costs nothing. Returns whether the line w
- `ask_people_panel` (says it can change data): Makes a panel, a named group of people the account owner invites, and returns its panel_id and a join link for your user to share. Each person joins on their own phone, states that they are an adult, is told what is recorded while they answer (how much played, whether the page stayed in front, pauses and replays, typed or pasted), chooses when they can be rung, and can leave at any time. Calls wit

## FAQ

### Is it free to use?

The README says the preview is free for simulated people and that no money is charged during the preview; every amount is test money. The site lists a 10% router fee and a per-person pay floor as a published price that is not yet charged.

### Can I try it without an account?

Yes. The README says the MCP server at api.person.tools/mcp can be used with simulated people and no key or account, and a test key can be obtained with POST /v1/keys.

### Are real strangers included?

Not during the preview. The site says strangers are not offered yet and requests for them are sent to the operator.

### Is the source code available?

Yes. The repository holds the service, the standard and the conformance suite under the Apache License 2.0.

### Is Ask real people: human feedback, user testing, surveys free?

Ask real people: human feedback, user testing, surveys has a free plan, with paid plans for more usage or features.

### How do I add Ask real people: human feedback, user testing, surveys to Claude Code?

Run this command in your terminal: claude mcp add --transport http human-feedback https://api.person.tools/mcp

### What are the best alternatives to Ask real people: human feedback, user testing, surveys?

The closest alternatives to Ask real people: human feedback, user testing, surveys are AgentExplorio, WritersLogic verifiable AI agents, MartNow Agent Exchange, MCP and Kariaa Agents.

## Where this comes from

- Description: written from the project's README, the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [AgentExplorio](https://ai-powered.com/mcp/agentexplorio): AgentExplorio is a free MCP server for agent tooling.
- [WritersLogic verifiable AI agents](https://ai-powered.com/mcp/writerslogic-verifiable-ai-agents): WritersLogic verifiable AI agents is a paid MCP server for agent tooling.
- [MartNow Agent Exchange](https://ai-powered.com/mcp/martnow-agent-exchange): MartNow Agent Exchange is an MCP server for agent tooling.
- [MCP](https://ai-powered.com/mcp/mcp-mwkad3hj): MCP is an open-source MCP server for agent tooling.
- [Kariaa Agents](https://ai-powered.com/mcp/kariaa-agents): Kariaa Agents is an open-source MCP server for agent tooling.
- [MCP Server](https://ai-powered.com/mcp/mcp-server-mwes9ff2): MCP Server is an open-source MCP server for agent tooling.
- [LiftMCP](https://ai-powered.com/mcp/liftmcp-2): LiftMCP is an MCP server for agent tooling.

---
Source: https://ai-powered.com/mcp/ask-real-people-human-feedback-user-testing-surveys
