Microsoft Planner Task Creator Server

A command-line tool and MCP (Model Context Protocol) server for creating and managing Microsoft Planner tasks. This project provides both a standalone Python CLI and a Node.js MCP server wrapper for AI assistant integration.

Created by: Babak Bandpey Website: cocode.dk

Website

• English
• فارسی (Persian)

Features

• 🔐 Secure OAuth Authentication: Device code flow with token caching
• 📋 Task Management: Create tasks with titles, descriptions, due dates, and labels
• 💬 Task Comments: Read and add comments to tasks via conversation threads
• 📝 Task Descriptions: Automatically included in all task listings and searches
• 🎯 Smart Resolution: Case-insensitive plan and bucket name resolution
• ⚙️ Flexible Configuration: CLI flags, environment variables, and config file support
• 🤖 MCP Integration: Expose Planner functionality to AI assistants like Claude
• 🧪 Comprehensive Testing: Full test suite with pytest

Architecture

┌─────────────────┐
│  AI Assistant   │
│ (Claude, etc.)  │
└────────┬────────┘
         │
         ▼
┌─────────────────┐
│  MCP Server     │
│  (Node.js/TS)   │
└────────┬────────┘
         │ spawns
         ▼
┌─────────────────┐
│  Python CLI     │
│  (planner.py)   │
└────────┬────────┘
         │
         ▼
┌─────────────────┐
│ Microsoft Graph │
│      API        │
└─────────────────┘

Installation

Prerequisites

• Python 3.8+ with pip
• Node.js 18+ with npm (for MCP server)
• Azure AD app registration with required permissions
• Access to Microsoft Planner

Azure AD Setup

💡 Don't have an Azure subscription? No problem! You don't need one. See SETUP_WITHOUT_AZURE_SUBSCRIPTION.md for free options including the Microsoft 365 Developer Program.

1. Register an app in Azure Portal
2. Set redirect URI to http://localhost (Public client/native)
3. Add API permissions:
   • Tasks.ReadWrite (required for all task operations)
   • Group.Read.All (required for reading task comments)
   • Group.ReadWrite.All (required for adding task comments)
   • offline_access (optional, for token refresh)
4. Grant admin consent for permissions (required for Group permissions)
5. Note your Tenant ID and Client ID

Note: The Group permissions require admin consent. After adding them, click "Grant admin consent" in the Azure Portal to enable comment functionality.

Quick Install (Automated)

Linux/macOS

# Clone the repository
git clone <repo-url>
cd planner-task-creator-cli-mcp

# Run the installation script
./scripts/INSTALLATION.sh

Windows (PowerShell)

# Clone the repository
git clone <repo-url>
cd planner-task-creator-cli-mcp

# Allow script execution (if needed)
Set-ExecutionPolicy -Scope CurrentUser RemoteSigned

# Run the installation script
.\scripts\INSTALLATION.ps1

Windows (Command Prompt)

REM Clone the repository
git clone <repo-url>
cd planner-task-creator-cli-mcp

REM Run the installation script
scripts\INSTALLATION.bat

Manual Install

Python CLI Setup (Linux/macOS)

# Clone the repository
git clone <repo-url>
cd planner-task-creator-cli-mcp

# Install Python dependencies
pip install -r requirements.txt

# Make CLI executable
chmod +x planner.py

# Optional: Create symlink for global access
mkdir -p ~/.planner-cli
cp planner.py ~/.planner-cli/
ln -s ~/.planner-cli/planner.py /usr/local/bin/planner

Python CLI Setup (Windows)

# Clone the repository
git clone <repo-url>
cd planner-task-creator-cli-mcp

# Create and activate virtual environment
python -m venv venv
.\venv\Scripts\Activate.ps1

# Install Python dependencies
pip install -r requirements.txt

# Optional: Copy CLI to user directory
mkdir "$env:USERPROFILE\.planner-cli" -Force
Copy-Item planner.py "$env:USERPROFILE\.planner-cli\"

MCP Server Setup

# Install Node dependencies
npm install

# Build TypeScript
npm run build

# The compiled server will be in dist/server.js

Configuration

Environment Variables

Linux/macOS (bash):

export TENANT_ID="your-tenant-id"
export CLIENT_ID="your-client-id"
export PLANNER_DEFAULT_PLAN="My Plan"
export PLANNER_DEFAULT_BUCKET="To Do"
export PLANNER_CONFIG_PATH="~/.planner-cli/config.json"  # Optional

Windows (PowerShell):

$env:TENANT_ID = "your-tenant-id"
$env:CLIENT_ID = "your-client-id"
$env:PLANNER_DEFAULT_PLAN = "My Plan"
$env:PLANNER_DEFAULT_BUCKET = "To Do"

Windows (Command Prompt):

set TENANT_ID=your-tenant-id
set CLIENT_ID=your-client-id
set PLANNER_DEFAULT_PLAN=My Plan
set PLANNER_DEFAULT_BUCKET=To Do

Config File

Create config file at:

• Linux/macOS: ~/.planner-cli/config.json
• Windows: %USERPROFILE%\.planner-cli\config.json

{
  "tenant_id": "your-tenant-id",
  "client_id": "your-client-id",
  "default_plan": "My Plan",
  "default_bucket": "To Do"
}

Configuration Precedence

1. CLI flags (highest priority)
2. Environment variables
3. Config file
4. Prompt for missing values (lowest priority)

Usage

Python CLI

Initialize Authentication

python planner.py init-auth

This will display a device code and URL for authentication. Visit the URL and enter the code to complete authentication.

Set Default Plan and Bucket

python planner.py set-defaults --plan "My Plan" --bucket "To Do"

List Plans

python planner.py list-plans

List Buckets

python planner.py list-buckets --plan "My Plan"

Create a Task

Minimal:

python planner.py add --title "Complete project report"

With all options:

python planner.py add \
  --title "Complete project report" \
  --plan "Q4 Projects" \
  --bucket "In Progress" \
  --desc "Write and submit quarterly report with metrics" \
  --due "2024-12-31" \
  --labels "Label1,Label3" \
  --verbose

MCP Server

Configuration

Add to your MCP client configuration (e.g., Claude Desktop):

{
  "mcpServers": {
    "planner": {
      "command": "node",
      "args": ["/path/to/planner-mcp-server/dist/server.js"],
      "env": {
        "TENANT_ID": "your-tenant-id",
        "CLIENT_ID": "your-client-id",
        "PLANNER_DEFAULT_PLAN": "My Plan",
        "PLANNER_DEFAULT_BUCKET": "To Do"
      }
    }
  }
}

Available Tools

Core Tools:

1. planner_initAuth: Initialize authentication
2. planner_createTask: Create a new task
3. planner_setDefaults: Set default plan and bucket
4. planner_listPlans: List available plans
5. planner_listBuckets: List buckets in a plan

Task Management: 6. planner_listTasks: List tasks in a plan or bucket (includes descriptions) 7. planner_findTask: Find a task by ID or title (includes description) 8. planner_completeTask: Mark a task as complete 9. planner_moveTask: Move a task to a different bucket 10. planner_updateTask: Update task properties (title, description, labels) 11. planner_deleteTask: Delete a task

Subtasks: 12. planner_addSubtask: Add a subtask (checklist item) 13. planner_listSubtasks: List subtasks for a task 14. planner_completeSubtask: Mark a subtask as complete

Comments: 15. planner_listComments: List all comments on a task 16. planner_addComment: Add a comment to a task

User Management: 17. planner_searchUsers: Search for users by name 18. planner_lookupUser: Resolve user identifier to full details

Bucket Management: 19. planner_createBucket: Create a new bucket 20. planner_deleteBucket: Delete a bucket 21. planner_renameBucket: Rename a bucket 22. planner_moveBucketTasks: Move all tasks from one bucket to another

API Reference

CLI Commands

init-auth

Initialize OAuth authentication with Microsoft.

Usage: python planner.py init-auth

set-defaults

Set default plan and bucket for task creation.

Options:

• --plan TEXT: Default plan name or ID (required)
• --bucket TEXT: Default bucket name or ID (required)

list-plans

List all available plans accessible to the user.

Usage: python planner.py list-plans

list-buckets

List all buckets in a specific plan.

Options:

• --plan TEXT: Plan name or ID (required)

add

Create a new task in Microsoft Planner.

Options:

• --title TEXT: Task title (required)
• --plan TEXT: Plan name or ID (optional if default is set)
• --bucket TEXT: Bucket name or ID (optional if default is set)
• --desc TEXT: Task description (optional)
• --due TEXT: Due date in YYYY-MM-DD format (optional)
• --labels TEXT: Comma-separated labels like "Label1,Label3" (optional)
• --verbose: Enable verbose output (optional)

list-tasks-cmd

List tasks in a plan or bucket. Task descriptions are automatically included.

Options:

• --plan TEXT: Plan name or ID (required)
• --bucket TEXT: Bucket name or ID (optional)
• --incomplete: Show only incomplete tasks (optional)

find-task-cmd

Find a task by ID or title. Returns full task details including description.

Options:

• --task TEXT: Task ID or title (required)
• --plan TEXT: Plan name or ID (required for title-based search)

list-comments-cmd

List all comments on a task.

Options:

• --task TEXT: Task ID or title (required)
• --plan TEXT: Plan name or ID (required)

add-comment-cmd

Add a comment to a task. Note: Comments can only be added to tasks that already have a conversation thread (typically created via Planner UI).

Options:

• --task TEXT: Task ID or title (required)
• --comment TEXT: Comment text to add (required)
• --plan TEXT: Plan name or ID (required)

Label Format

Labels should be specified as comma-separated values: Label1,Label2,Label3

These are mapped to Planner categories:

• Label1 → category1
• Label2 → category2
• etc.

Development

Running Tests

# Run all tests
pytest

# Run with coverage
pytest --cov=planner --cov-report=html

# Run specific test file
pytest tests/test_auth.py

# Run with verbose output
pytest -v

Project Structure

.
├── planner.py              # Main Python CLI
├── src/
│   └── server.ts          # MCP server implementation
├── tests/
│   ├── conftest.py        # Test fixtures
│   ├── test_auth.py       # Authentication tests
│   ├── test_config.py     # Configuration tests
│   ├── test_resolution.py # Resolution tests
│   ├── test_task_creation.py # Task creation tests
│   └── test_cli_commands.py  # CLI command tests
├── 0-docs/
│   ├── prd.md             # Product requirements
│   └── implementation steps/  # Detailed specs
├── requirements.txt        # Python dependencies
├── package.json           # Node.js dependencies
└── README.md              # This file

Module Overview

1. Authentication (001): OAuth device code flow with MSAL
2. Graph Client (002): HTTP client for Microsoft Graph API
3. Configuration (003): Config file management
4. Resolution (004): Plan and bucket name-to-ID resolution
5. Task Creation (005): Task creation with all fields
6. CLI Commands (006): Typer-based CLI interface
7. Error Handling (007): Structured JSON error responses
8. MCP Server (008): Node.js wrapper for AI integration
9. Testing (009): Comprehensive test suite

Error Handling

All errors are returned as structured JSON:

{
  "code": "ErrorCode",
  "message": "Human-readable error message",
  "candidates": [{"id": "...", "name": "..."}]
}

Error Codes

• ConfigError: Missing required configuration
• NotFound: Resource not found (with candidates)
• Ambiguous: Multiple matches (with candidates)
• AuthError: Authentication failure
• UpstreamError: Graph API error

Security

• 🔐 Tokens are stored securely:
  • Linux/macOS: ~/.planner-cli/msal_cache.bin with 0600 permissions
  • Windows: %USERPROFILE%\.planner-cli\msal_cache.bin
• 🔒 Config file permissions are set to 0600 (Unix) or user-only access (Windows)
• 🚫 Tokens are never logged or exposed in output
• ✅ OAuth device code flow for secure authentication

Troubleshooting

Authentication Issues

Problem: Device code flow times out

Solution: Ensure you complete authentication within 15 minutes and have the required permissions

Resolution Issues

Problem: "Plan not found" error

Solution: Use list-plans to see available plans. Plan names are case-insensitive but must match exactly.

Problem: "Multiple plans match" (Ambiguous)

Solution: Use the plan ID instead of name, or ensure unique naming

Permission Issues

Problem: "Authorization failed: insufficient permissions"

Solution: Ensure your Azure AD app has the required API permissions and admin consent

MCP Server Issues

Problem: MCP server not connecting

Solution:

1. Ensure Python CLI is working standalone first
2. Check that PLANNER_CLI_PATH points to correct location
3. Verify environment variables are set in MCP config

Project Structure

planner-task-creator-cli-mcp/
├── planner.py              # Main CLI entry point
├── planner_lib/            # Modular Python library (25 files)
│   ├── auth.py            # Authentication
│   ├── config.py          # Configuration
│   ├── graph_client.py    # Graph API client
│   ├── resolution*.py     # Plan/bucket resolution
│   ├── task_*.py          # Task operations
│   └── cli_*.py           # CLI commands
├── src/                   # TypeScript MCP server
│   ├── server.ts          # Main entry point
│   └── server/            # Modular server (7 files)
├── tests/                 # Test suite (68 tests)
│   ├── test_*.py          # Unit tests
│   └── test_task_*/       # Modular test suites
├── docs/                  # Documentation
│   ├── README.md          # Documentation index
│   ├── QUICKSTART.md      # Quick start guide
│   ├── SETUP_GUIDE.md     # Complete setup
│   └── *.md               # Other guides
├── scripts/               # Installation & setup scripts
│   ├── INSTALLATION.sh    # Automated installer (Linux/macOS)
│   ├── INSTALLATION.ps1   # Automated installer (Windows PowerShell)
│   ├── INSTALLATION.bat   # Automated installer (Windows CMD)
│   ├── setup-cursor-mcp.sh # Cursor setup
│   └── test-*.sh          # Test scripts
├── 0-docs/                # Implementation specs
│   ├── implementation steps/ # Detailed specs
│   ├── prd.md            # Product requirements
│   └── ROADMAP.md        # Project roadmap
├── package.json           # Node.js dependencies
├── requirements.txt       # Python dependencies
└── README.md              # This file

Contributing

Contributions are welcome! Please:

1. Follow the existing code structure
2. Add tests for new features
3. Update documentation
4. Ensure all tests pass

License

MIT License - see LICENSE file for details

Support

For issues and questions:

• Check the troubleshooting section
• Review the detailed specifications in 0-docs/implementation steps/
• Open an issue on GitHub

Author

Babak Bandpey — cocode.dk | LinkedIn | GitHub

License

Apache-2.0 | © 2026 Cocode | Created by Babak Bandpey

Acknowledgments

• Microsoft Graph API for Planner integration
• MSAL Python library for authentication
• Typer and Rich for CLI framework
• MCP SDK for AI assistant integration