AI Task Manager for Claude Workflows

Overview

The AI Task Manager for Claude Workflows is a Model Context Protocol (MCP) server designed to manage and structure multi-step tasks for AI assistants like Claude. It provides a structured task queue, progress tracking, and user approval checkpoints to ensure efficient and organized workflows.

Features

• Task Planning: Create and manage tasks with multiple steps.
• Progress Tracking: Monitor the status of tasks and projects.
• User Approval: Require user approval for completed tasks and projects.
• Task Visualization: View detailed task information.
• State Management: Enforce task status transition rules.
• Enhanced CLI: Inspect and manage tasks via command-line interface.

Configuration

To configure the tool in your MCP client (e.g., Claude Desktop or Cursor), use the following JSON configuration:

{
  "tools": {
    "taskqueue": {
      "command": "npx",
      "args": ["-y", "taskqueue-mcp"]
    }
  }
}

Usage

CLI Commands

To view available commands and options, use:

npx task-manager-cli --help

Task Approval

Tasks must be marked as "done" before approval. Use the following CLI command:

npm run approve-task -- <projectId> <taskId>

Listing Tasks and Projects

List all projects and tasks:

npm run list-tasks

View details of a specific project:

npm run list-tasks -- -p <projectId>

Available MCP Tools

Project Management Tools

• list_projects: Lists all projects.
• read_project: Gets details of a specific project.
• create_project: Creates a new project with initial tasks.
• delete_project: Removes a project.
• add_tasks_to_project: Adds new tasks to an existing project.
• finalize_project: Finalizes a project after all tasks are done.

Task Management Tools

• list_tasks: Lists all tasks for a specific project.
• read_task: Gets details of a specific task.
• create_task: Creates a new task in a project.
• update_task: Modifies a task's properties.
• delete_task: Removes a task from a project.
• approve_task: Approves a completed task.
• get_next_task: Gets the next pending task in a project.
• mark_task_done: Marks a task as completed with details.

Task Status and Workflows

Tasks follow a specific workflow with defined status transitions:
- Not Started → In Progress
- In Progress → Done or Not Started
- Done → In Progress (if additional work is needed)

Data Schema and Storage

File Location

Task data is stored in a JSON file at the following default locations:
- Linux: ~/.local/share/taskqueue-mcp/tasks.json
- macOS: ~/Library/Application Support/taskqueue-mcp/tasks.json
- Windows: %APPDATA%\taskqueue-mcp\tasks.json

Data Schema

The JSON file follows this structure:

TaskManagerFile
├── projects: Project[]
    ├── projectId: string
    ├── initialPrompt: string
    ├── projectPlan: string
    ├── completed: boolean
    └── tasks: Task[]
        ├── id: string
        ├── title: string
        ├── description: string
        ├── status: string
        ├── approved: boolean
        ├── completedDetails: string
        ├── toolRecommendations: string
        └── ruleRecommendations: string

License

This project is licensed under the MIT License.

About

The AI Task Manager for Claude Workflows is an MCP tool designed to guide AI agent workflows, particularly for managing the enthusiasm of Claude. It provides a structured approach to task management, ensuring clarity and efficiency in AI-driven projects.