mcp-task-manager

The MCP Task Manager Server provides a persistent, local backend for managing projects and tasks via MCP-compatible tools. It stores data in a local SQLite database and exposes a suite of tools for creating projects, adding and updating tasks, exporting/importing project data, and querying task state. Clients (such as AI agents or scripts) can interact with these tools to manage structured task data within distinct projects. The design emphasizes client-driven workflows, enabling clients to orchestrate complex task hierarchies while the server focuses on reliable data storage and standard tool interfaces.

Key capabilities include creating a new project, adding tasks with optional dependencies and priorities, listing or showing tasks with optional subtasks, expanding tasks into subtasks, updating task statuses, and exporting/importing entire projects as JSON for backup or transfer. The tooling set is designed to be used by MCP-enabled clients to build automation around project timelines, task dependencies, and prioritization, without imposing a fixed workflow on the client.

Prerequisites:

• Node.js (LTS) and npm installed on your system
• Git (optional, for cloning the repository)

Install from source:

1. Clone the repository git clone https://github.com/bsmi021-mcp-task-manager-server.git cd bsmi021-mcp-task-manager-server

2. Install dependencies npm install

3. Optional: Configure environment variables (see below) and create data directory if needed mkdir -p ./data

4. Run in development mode (uses ts-node and nodemon for auto-reloading) npm run dev

5. Build for production npm run build

6. Run production server npm start

Prerequisites for production run:

• A Node.js runtime (as above)
• A properly configured environment with the required database path and log level (see Environment Variables).

Environment variables:

• DATABASE_PATH: Path to the SQLite database file. Default is ./data/taskmanager.db.
• LOG_LEVEL: Logging level. Common values: debug, info, warn, error. Default is info.

Tips and common issues:

• Ensure the data directory exists or is creatable by the process to avoid database file creation errors.
• If the server fails to connect to the database, verify file permissions on the ./data directory.
• When exporting a project, you can store the JSON string externally and use importProject to recreate a new project with the same data structure elsewhere.
• For long-running deployments, monitor logs for any SQLite file locking or disk space issues.
• The server connects via stdio in development mode; in production you’ll typically run the built dist/server.js as a standalone process.