gotoHuman MCP Server

Official

Add human approvals to AI agents and agentic workflows

How to install ↓ Compare GitHub source repository ↗

Category: Collaboration #35 of 55
Stars: ★ 52 Popular
Transport: stdio (local process)
Runtime: Node.js
Credentials: API key / credential required
License: MIT
Last commit: 10 days ago
Tools: 3

50FMRS · D

The gotoHuman MCP server provides a secure, controlled way to incorporate human approvals into AI workflows, simplifying the oversight of AI agents.

Strongest · Maintenance 11/20 Weakest · Reliability 9/20

Reliability

9/20

Security and permissions

10/20

Maintenance

11/20

Documentation

10/20

Setup experience

10/20

Why each score

Reliability 9/20: Source suggests a standard TypeScript structure (tsconfig.json, package.json, src/ directory), but tool names in README may not exactly match actual declarations in src files (e.g., README mentions 'request-human-review-with-form' while source may use 'requestHumanReview'). This inconsistency reduces reliability as we cannot confidently infer actual behavior. No tests or CI (no .github directory) are provided, so key paths are unverified. The handshake is simple but lack of committed tests caps reliability at 12.
Security and permissions 10/20: Credential handling is via GOTOHUMAN_API_KEY environment variable, which is good practice. The server makes external API calls, which is expected since it's a managed service. No malicious activity detected. Write operations, like creating tickets, have side effects without confirmation, but this is inherent to the service. However, least privilege is not explicitly articulated, and there is no built-in confirmation for potentially dangerous operations within MCP. Nonetheless, no dangerous defaults are present, so score remains at 10.
Maintenance 11/20: The repository has recent commits, an MIT license, and stars (53). There is no CI or automated tests, which is a gap in continuous integration. Maintenance activity is visible, but there is no clear evidence of versioning or dependency updates. The score reflects active maintenance with governance gaps.
Documentation 10/20: README provides clear installation instructions, including usage examples and tool descriptions. However, it lacks detailed parameter types and examples. Limitations, costs, and troubleshooting are not explicitly addressed. Documentation is usable but could be more comprehensive.
Setup experience 10/20: Setup is relatively straightforward: run the package via npx and provide environment variables. Examples for multiple clients (Cursor, Claude, Windsurf) including Windows are given. However, there is no evidence of execution, so there could be subtle issues with the build or environment that are not covered.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkChanges third-party account data

Best for

Developers building AI agents that require human supervision
Teams wanting to implement approval processes in AI workflows
Scenarios needing a customizable approval UI

Not for

Fully automated processes without human intervention
Offline or intranet environments (requires network connectivity)
On-premise deployments without the hosted service

Required permissions

Requires a GOTOHUMAN_API_KEY for authentication and access to forms in your account
Requires network access to app.gotohuman.com
Write operations involve the gotoHuman service, no direct local file access

Risks and side effects

Exposure of API key could lead to unauthorized approval requests
Network issues may disrupt communication with the service
Over-reliance on the service could cause delays in approvals

Setup

Before you start

Runtime：Node.js

GOTOHUMAN_API_KEY requiredsecret API key for your gotoHuman account, obtained at app.gotohuman.com; when running locally you also set GOTOHUMAN_AGENT_ID (your-agent-id).

Other optional settings (1)

GOTOHUMAN_AGENT_ID optional

Install using npx: npx @gotohuman/mcp-server. Configure your MCP client (e.g., Cursor, Claude, Windsurf) by adding an environment variable GOTOHUMAN_API_KEY with your API key. For local development, install dependencies with npm install, build with npm run build, and run as a local server using node /path/to/build/index.js.

claude_desktop_config.json

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

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json

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

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal

claude mcp add gotoHuman -e GOTOHUMAN_API_KEY=your-api-key -- npx -y @gotohuman/mcp-server

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

The client's tool list should show list-forms, get-form-schema, and request-human-review-with-form; calling list-forms and getting back your review templates confirms the connection.

Troubleshooting

Ensure your API key is correct and active.
Check network connectivity to app.gotohuman.com.
If using a local build, confirm the build succeeded and the path is correct.
If tools fail, verify form IDs exist and fetch the correct schema using get-form-schema.

Things to try

Once connected, you can ask your AI assistant things like:

List all available review templates in my gotoHuman account
Fetch the form schema for a review template so I can request a human review
Submit a human review request with this template and notify me via webhook when it's reviewed
Assign this review request to a specific user email on my team

Tools 3

list-forms read-only

List all available review templates in your account, including high-level info about added fields.

get-form-schema read-only

Get the schema to use when requesting a human review for a given review template.

request-human-review-with-form writes

Request a human review. Will appear in your gotoHuman inbox.

Use cases

Request human approval before AI agents execute critical actions

Review AI-generated outputs in automated workflows

Implement staged human approvals in multi-step agentic workflows

Integrate human oversight into AI-powered coding assistants

Supported clients

Claude Desktop

Cursor

Windsurf

Listed from the project's documentation, not tested by this site.

Overview

The gotoHuman MCP server enables AI agents and agentic workflows to request human approvals via MCP. It provides a fully managed, async human-in-the-loop workflow with a customizable approval UI, built-in auth, webhooks, notifications, and team features. This server acts as a bridge between AI systems and human reviewers, enabling human oversight of AI outputs.

Similar servers

Atlassian Rovo MCP Server 66 · C

Securely connect Atlassian Cloud to AI tools.

★ 1.1k Compare with this →

Notion MCP Server 64 · C

Official Notion MCP server for querying, editing, and managing Notion pages and databases via AI agents.

★ 4.7k · Tools 9 Compare with this →

WhatsApp Channel for Claude Code 64 · C

Drive your Claude Code session straight from WhatsApp

★ 103 · Tools 10 Compare with this →

Anytype MCP Server 64 · C

Official MCP server that lets AI assistants manage your encrypted Anytype wiki in natural language.

★ 531 Compare with this →

Compare with other MCP servers

Source revision fda85809bb94 Data synced 2026-10-11 Read the FMRS scoring method