1WLF TMS — MCP Server

Connect AI assistants to the 1WLF Task Management System. Manage organizations, projects, and tasks through natural language via Claude Code, Claude Desktop, or any MCP-compatible client.

Quick Start

Claude Code

claude mcp add tms -- uvx 1wlf-tms-mcp

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "tms": {
      "command": "uvx",
      "args": ["1wlf-tms-mcp"]
    }
  }
}

Requires uv to be installed. uvx handles downloading and running the server automatically — no manual Python setup needed.

pip

pip install 1wlf-tms-mcp
tms-mcp  # runs the stdio server

Usage

Start a session and authenticate first:

"Log into TMS as user@example.com"

Then interact naturally:

"What organizations do I belong to?"

"List all blocked tasks in project 5"

"Create a high-priority bug 'Fix login redirect' in project 3"

"Show me task 42 with its audit history"

"Move task 42 to done and assign task 15 to user 7"

Tools

Task Fields

Self-Hosted

Override the API URL for your own TMS instance:

{
  "mcpServers": {
    "tms": {
      "command": "uvx",
      "args": ["1wlf-tms-mcp"],
      "env": {
        "TMS_API_URL": "https://your-tms-api.example.com"
      }
    }
  }
}

Requirements

• Python 3.10+
• A 1WLF TMS account

Links

• TMS Web App
• MCP Protocol