Gotohuman

Provides a managed MCP server to request human approvals from AI workflows with an approval UI and webhooks.
  • javascript

52

GitHub Stars

javascript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "gotohuman-gotohuman-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@gotohuman/mcp-server"
      ],
      "env": {
        "GOTOHUMAN_API_KEY": "your-api-key"
      }
    }
  }
}

The gotoHuman MCP Server lets you request human approvals from AI workflows inside your tooling, with an async, fully managed human-in-the-loop experience, built-in authentication, webhooks, notifications, and team features. It enables you to integrate a customizable approval UI into your AI pipelines so that decisions requiring human review can be handled efficiently and securely.

How to use

You can connect your AI workflows to gotoHuman MCP Server from your MCP client (such as Cursor, Claude, or Windsurf). Use a local or remote MCP config to point your client at the server so that AI outputs requiring human review are routed to your gotoHuman inbox. You will typically configure a server entry with an executable that runs the MCP server, and supply your API key for authentication. When a review is needed, the system will present the form in the approval UI, and you or your team can approve or reject the request from within your workflow.

Two common runtime configurations are provided: one runs the MCP server via a package runner, and another runs a built local server from a compiled build. The first is suitable for quick trials, while the second is used when you are developing or hosting a local build of the server.

Configuring the MCP client to use gotoHuman

{
  "mcpServers": {
    "gotoHuman": {
      "command": "npx",
      "args": ["-y", "@gotohuman/mcp-server"],
      "env": {
        "GOTOHUMAN_API_KEY": "your-api-key"
      }
    }
  }
}

Run a local build with MCP client

If you are developing locally, you can start from a built index and run the server via a node command. This is useful when you have a local build you want to test inside an MCP client.

{
  "mcpServers": {
    "gotoHuman": {
      "command": "node",
      "args": ["/<absolute-path>/build/index.js"],
      "env": {
        "GOTOHUMAN_API_KEY": "your-api-key"
      }
    }
  }
}

Available tools

list-forms

List all available review forms and associated high-level field information.

get-form-schema

Fetch the schema for a specific form so you know which fields to populate when requesting a human review.

request-human-review-with-form

Submit a human review request using a specific form ID and field data, optionally assigning the task to particular users and including metadata for the webhook response.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational