SuperAPI AI Directory
Partner with us中文AI newsFavorites
All toolsAI agentsgotohuman-mcp-server

gotohuman-mcp-server

Human-in-the-loop platform - Allow AI agents and automations to send requests for approval to your gotoHuman inbox.

Open website

About this tool

gotoHuman MCP Server is the official MCP (Model Context Protocol) server provided by gotoHuman. Its goal is to enable AI agents and agent workflows to easily add human approval (human-in-the-loop) steps.

Purpose

According to the repository description, gotoHuman is designed to add human approval to AI agents and agent workflows. It provides a fully managed asynchronous human-in-the-loop workflow, equipped with a customizable approval UI, and built-in authentication, webhooks, notifications, team features, and a continuously evolving training dataset. This MCP server is used to request human approval from AI workflows or to assist with integration.

Main Capabilities

This MCP server exposes the following tools (functions):

  1. List all available review types: Returns all available review types in the account, including high-level information about the added fields.

  2. get-form-schema: Retrieves the schema needed to request human review for a specified review type.

    • Parameter formId: The review type ID for which to get the schema.
    • Return value: The schema after considering the included fields and their configuration.

  3. request-human-review-with-form: Requests a human review, which will appear in the gotoHuman inbox.

    • Parameter formId: The ID of the review type to use.
    • Parameter fieldData: The content (AI output to be reviewed, context, etc.) and the configuration of the fields for that review type. Its schema needs to be obtained via get-form-schema.
    • Parameter config (optional): The configuration for that review type. Its schema also needs to be obtained via get-form-schema.
    • Parameter title (optional): The title displayed in the inbox and notifications.
    • Parameter webhookUrl (optional): The webhook URL used for this request (a static URL can be set on the review type or agent in gotoHuman).
    • Parameter workflow (optional): An object that associates this review with a multi-step agent workflow:
      • runId: A unique ID for the current workflow run, used to associate multiple steps.
      • metadata: Optional additional data that will be included in the webhook response after the review template is submitted.
    • Parameter assignToUsers (optional): A list of specified user emails to assign the review to.
    • Return value: reviewId and reviewLink.

Usage

According to the repository description, this MCP server is available via npm (the README mentions "Available on npm as:", but does not provide a specific package name). The setup method is:

  1. Get an API key at app.gotohuman.com and set up an approval step.
  2. Configure it in environments such as Cursor / Claude.

The README also provides a Windows-related path hint: the args path needs to be written as C:\\<absolute-path>\\build\\index.js.

The repository itself is open source and located at https://github.com/gotohuman/gotohuman-mcp-server , and you can participate in development by registering a GitHub account.

Applicable Scenarios

Suitable for teams that need to let AI agents or automated processes pause at key points and wait for human review or approval. For example, having AI-generated content confirmed by a human, building multi-step agent workflows and inserting approval steps into them, etc.

Limitations

The provided materials do not specify the server's specific price, charging model, platform compatibility scope, performance metrics, or effectiveness data. This information is not currently available.

Related tools

Browse category