Documentation — My Fordyce

My Fordyce Documentation

Learn how to integrate your AI coding assistant with the My Fordyce project management platform using the Model Context Protocol (MCP).

1. Quickstart Guide

Getting your AI assistant connected to your project board takes less than 5 minutes:

Sign up for an account at myfordyce.fordycecg.com.
Create a new Project.
Navigate to Settings > API Keys and generate a new long-lived key.
Add the project-fordyce MCP server to your AI client's configuration file (see below).
Start chatting with your AI to manage your project!

2. MCP Server Configuration

The My Fordyce MCP server runs via npx and communicates over standard I/O (stdio). You will need Node.js installed to run the server.

Claude Desktop

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "project-fordyce": {
      "command": "npx",
      "args": ["-y", "@fordycecg/mcp-server-project-fordyce"],
      "env": {
        "FORDYCE_API_KEY": "<your-api-key>",
        "FORDYCE_PROJECT_ID": "<your-project-id>"
      }
    }
  }
}

Gemini / API Usage

If you are injecting the MCP server via the core SDK or local tools like mcp-cli, structure the environment variables similarly before executing the npx wrapper.

3. Available MCP Tools

Once connected, your AI assistant will automatically have access to the following project management tools:

Tool Name	Description
`list_tasks`	Retrieves the current kanban board backlog.
`create_task`	Logs a new bug, feature, or idea ticket.
`update_task`	Modifies task metadata (e.g., adding remediation details).
`move_task`	Moves a task between statuses (e.g., to 'in_progress').
`list_roadmap_features`	Retrieves high-level strategic roadmap goals.

Workflow Tip

When your AI assistant finishes a task, ask it to "close the task and log the remediation details". It will update the ticket description with the exact files it modified before moving it to Done!