ado-task-manager

AdoTaskManager

An MCP server to manage Task work items in Azure DevOps.

Requirements

• Python 3.10+
• Azure DevOps Personal Access Token with Work Items (read/write) access

Installation

Local development

python3 -m venv .venv
source .venv/bin/activate
pip install -e .
cp .env.example .env
# edit .env with your AZDO_ORG_URL, AZDO_PAT, AZDO_PROJECT

Alternatively, if you prefer uv:

uv venv
source .venv/bin/activate
uv pip install -e .

Running ado-task-manager --help will confirm the CLI is available.

ado-task-manager --help

CLI usage

source .venv/bin/activate
# Fetch a task
ado-task-manager fetch --id 123

# Fetch my tasks (active only)
ado-task-manager mine

# Fetch my tasks (including closed)
ado-task-manager mine --closed

# Fetch child tasks
ado-task-manager children --id 123

# Create a task
ado-task-manager create --title "New Task" --parent-id 123 --description "Description"

MCP server

Installing via mcp

mcp

The official MCP tooling expects uv to be available:

uv

uv tool install mcp
# Register the server with an MCP client (e.g., Claude Desktop)
uv run mcp install src/ado_task_manager/server.py --name "ADO Task Manager"
# Launch the MCP Inspector for interactive testing
uv run mcp dev src/ado_task_manager/server.py --with-editable .

The install command accepts -v KEY=value or -f .env flags if you want to supply Azure DevOps environment variables during registration.

-v KEY=value

-f .env

Manual MCP client registration

You can use the server command to run the MCP server directly. This is useful when installing via PyPI.

{
  "mcpServers": {
    "ado-task-manager": {
      "command": "ado-task-manager",
      "args": [
        "server"
      ],
      "env": {
        "AZDO_ORG_URL": "https://dev.azure.com/<org>",
        "AZDO_PAT": "<personal_access_token>",
        "AZDO_PROJECT": "<default_project>"
      }
    }
  }
}

Merge this snippet with existing definitions as needed, then restart or refresh your client so it discovers the server.