MCP Integration Guide

Learn how to connect Workunit to your development tools using the Model Context Protocol (MCP)

Last updated: January 2025

Overview

The Model Context Protocol (MCP) allows AI development tools to access Workunit directly. Once connected, AI assistants can create workunits, manage tasks, link assets, and track your development work—all without leaving your editor or chat interface.

This guide shows you how to add the Workunit MCP server to your AI tools, starting with Claude Code.

Prerequisites

  • • A Workunit account (sign up at workunit.app)
  • • Claude Code installed on your system
  • • An active internet connection

Claude Code

Adding Workunit to Claude Code is a single command. Once installed, Claude Code can manage your workunits, tasks, and assets directly within your development workflow.

Installation

Run this command in your terminal:

claude mcp add --transport http workunit https://workunit.app/mcp

This command registers the Workunit MCP server with Claude Code using HTTP transport.

Verify Installation

After installation, verify that Workunit is connected:

  1. Open Claude Code
  2. Start a new conversation
  3. Ask: "What Workunit tools do you have access to?"
  4. Claude should list the available Workunit MCP tools

Available Tools

Once connected, Claude Code has access to these Workunit capabilities:

  • • Create & manage workunits: Define problems, set success criteria, track status
  • • Task management: Create tasks, assign work, track progress
  • • Asset tracking: Link products, systems, people, and knowledge to your work
  • • Context preservation: Access and update workunit context and AI-generated insights

Troubleshooting

Connection Failed

If Claude Code can't connect to Workunit, verify your internet connection and ensure https://workunit.app is accessible.

Tools Not Available

Restart Claude Code after installation. If tools still don't appear, try removing and re-adding the MCP server.

Authentication Issues

The first time Claude Code uses Workunit tools, you'll be prompted to authenticate. Follow the OAuth flow to grant access.

Gemini CLI

Gemini CLI is Google's command-line AI assistant. Connect it to Workunit to manage your workunits, tasks, and assets directly from your terminal.

Installation

Run this command in your terminal:

gemini mcp add --transport http workunit https://workunit.app/mcp

This command registers the Workunit MCP server with Gemini CLI using HTTP transport.

Authentication

After installation, you need to authorize Gemini CLI to access your Workunit account:

  1. Launch Gemini CLI in your terminal
  2. Type the authentication command: /mcp auth workunit
  3. You'll be redirected to the OAuth authorization screen in your browser
  4. Review the permissions and click "Authorize" to grant access
  5. Return to your terminal - authentication is complete!

Verify Installation

Test that Workunit is connected and working:

  1. In Gemini CLI, type: "Ping workunit MCP server"
  2. You should receive a confirmation message that the server is responding
  3. Try asking: "What Workunit tools are available?"
  4. Gemini will list all available Workunit MCP tools

Available Tools

Once connected, Gemini CLI has access to these Workunit capabilities:

  • • Create & manage workunits: Define problems, set success criteria, track status
  • • Task management: Create tasks, assign work, track progress
  • • Asset tracking: Link products, systems, people, and knowledge to your work
  • • Context preservation: Access and update workunit context and AI-generated insights

Troubleshooting

Connection Failed

If Gemini CLI can't connect to Workunit, verify your internet connection and ensure https://workunit.app is accessible.

Authentication Failed

If the /mcp auth workunit command fails, try restarting Gemini CLI and running the command again. Make sure you complete the OAuth flow in your browser.

Tools Not Available

If Workunit tools don't appear after authentication, restart Gemini CLI. If the issue persists, try removing and re-adding the MCP server.

Other Tools

Workunit currently supports Claude Code and Gemini CLI (see above for setup instructions). Support for additional MCP-compatible tools is coming soon:

  • • Claude Desktop (coming soon)
  • • Cline (coming soon)
  • • Other MCP-compatible tools (coming soon)