MCP.so
ログイン
D

Datapoint

@impel-intelligence

リポジトリ

Datapoint について

MCP server for human-in-the-loop surveys, A/B preference tests, ratings, and rankings. Get real human feedback inside Claude Code, Claude Desktop, Cursor, Windsurf, and any MCP client

#data-labelling#human-surveys#human-judgement#a-b-test#real-human-opinions

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "datapoint-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/impel-intelligence/datapoint-mcp.git",
        "datapoint-mcp"
      ]
    }
  }
}

ツール

11

Authenticate with your Datapoint AI account (opens browser)

Upload local images, audio, or video so they can be used in a survey

Design a survey from a natural language description

Launch a survey from a plan

Check status, progress, and aggregated results

Get raw per-annotator responses (paginated)

List all your surveys

Pause task serving for an active survey (in-flight responses keep arriving)

Resume task serving for a paused survey

Check your account balance

Open a checkout link to top up your account

概要

What is Datapoint?

Datapoint is an MCP server that enables Claude, GPT, Gemini, and other MCP-capable agents to recruit real humans for evaluation tasks. It lets you run surveys, A/B preference comparisons, ratings, and rankings on text, images, audio, and video directly from your editor, then return aggregated results into the conversation.

How to use Datapoint?

Install via a plugin (e.g., /plugin marketplace add impel-intelligence/datapoint-mcp in Claude Code) or add the MCP server configuration to your client (using uvx). First run the setup tool to authenticate with your Datapoint AI account. Then simply ask your agent to design and launch a survey, monitor progress with check_survey, and retrieve results.

Key features of Datapoint?

  • Authenticate via browser-based OAuth or API key (DATAPOINT_API_KEY)
  • Upload local images, audio, and video for use in surveys
  • Design surveys from natural language descriptions (plan_survey)
  • Launch surveys, check status, and get per-annotator responses
  • Supports chain surveys with up to 5 dependent steps and skip conditions
  • Pay-as-you-go pricing with balance check and credit top-up (add_credits)

Use cases of Datapoint?

  • A/B test logos, landing pages, ad creative, and copy (design & UX)
  • Human ratings of model outputs for AI evaluation and hallucination checks
  • Collect RLHF/DPO preference pairs at scale
  • Dataset labeling including classification, ranking, captioning, and content moderation
  • Quick product research like concept tests, naming, and pricing reads

FAQ from Datapoint

How do I authenticate with Datapoint?

Run the setup tool to open a browser for authentication. Alternatively, set the DATAPOINT_API_KEY environment variable to override saved config.

What tools does Datapoint provide?

The server offers 11 tools: setup, upload_media, plan_survey, create_survey, check_survey, get_survey_responses, list_surveys, pause_survey, resume_survey, check_balance, and add_credits.

Can I run chain surveys with dependent questions?

Yes. You can describe a multi-step survey where later questions depend on earlier answers. The plan_survey tool designs the chain with skip conditions, and costs are shown as an upper bound.

How long does it take to get responses?

Datapoint returns responses in minutes, compared to minutes-to-hours for Mechanical Turk and Prolific.

What media types are supported?

You can upload images, audio, and video via upload_media, and include them in surveys alongside text.

よくある質問

How do I authenticate with Datapoint?

Run the `setup` tool to open a browser for authentication. Alternatively, set the `DATAPOINT_API_KEY` environment variable to override saved config.

What tools does Datapoint provide?

The server offers 11 tools: `setup`, `upload_media`, `plan_survey`, `create_survey`, `check_survey`, `get_survey_responses`, `list_surveys`, `pause_survey`, `resume_survey`, `check_balance`, and `add_credits`.

Can I run chain surveys with dependent questions?

Yes. You can describe a multi-step survey where later questions depend on earlier answers. The `plan_survey` tool designs the chain with skip conditions, and costs are shown as an upper bound.

How long does it take to get responses?

Datapoint returns responses in minutes, compared to minutes-to-hours for Mechanical Turk and Prolific.

What media types are supported?

You can upload images, audio, and video via `upload_media`, and include them in surveys alongside text.

コメント

基本情報

カテゴリ

開発者ツール

トランスポート

stdio

公開者

impel-intelligence

投稿者

Akshat Jagga

「開発者ツール」の他のコンテンツ

Deepwiki MCP Server

regenrek

📖 MCP server for fetch deepwiki.com and get latest knowledge in Cursor and other Code Editors

1.4K

Code Index MCP

johnhuang316

A Model Context Protocol (MCP) server that helps large language models index, search, and analyze code repositories with minimal setup

979

MCP Unity Editor (Game Engine)

CoderGamester

Model Context Protocol (MCP) plugin to connect with Unity Editor — designed for Cursor, Claude Code, Codex, Windsurf and other IDEs

1.8K

MCP Containers

metorial

Connect any AI model to 1200+ integrations (MCP, CLI, API)

3.3K

mcp-excalidraw

yctimlin

MCP server and Claude Code skill for Excalidraw — programmatic canvas toolkit to create, edit, and export diagrams via AI agents with real-time canvas sync.

2.1K