Tama - AI-Powered Task Manager CLI ✨

What is Tama?

Tama is a command-line interface (CLI) tool for task management, enhanced with AI capabilities for task generation and decomposition. It uses an OpenAI-compatible API configured for DeepSeek models to parse product requirements documents (PRDs) and break down complex tasks into manageable subtasks.

How to use Tama?

Install Tama with pip in a Python 3.12 virtual environment, then configure a .env file with your DeepSeek API key and model settings. Run commands like tama add, tama list, tama expand, and tama prd in the terminal. Tama can also be used as an MCP server by running uv --directory /path/to/TAMA_MCP run python -m src.mcp_server.

Key features of Tama

• Standard task management with dependency tracking
• AI-powered PRD parsing and task decomposition
• SQLite database for persistent data storage
• Git integration for automatic branch and commit creation
• Code stub generation from task details
• Dependency cycle detection and visualization
• Markdown and Mermaid report generation

Use cases of Tama

• Parse a product requirements document to automatically generate a structured task list
• Break down high-level tasks into detailed subtasks using AI
• Track dependencies between tasks and detect circular dependencies
• Generate placeholder code files based on task descriptions
• Identify the next actionable task based on status and dependencies

FAQ from Tama

What dependencies or runtime does Tama require?

Tama requires Python 3.12 and uses a virtual environment. It depends on the tama-cli package, SQLite for storage, and an OpenAI-compatible API key (DeepSeek models are configured by default).

How does Tama handle data persistence?

All task data is stored in a local SQLite database, ensuring stable and reliable data operations.

Can Tama be used programmatically by other applications?

Yes, Tama can be run as an MCP server over stdio transport, exposing tools like list_tasks, add_task, and set_status for external integration.

Does Tama support any shell auto‑completion?

Yes, you can set up shell auto‑completion by running tama --install-completion.

What AI models does Tama use?

Tama is configured to use DeepSeek models through an OpenAI-compatible API. The configuration requires a DEEPSEEK_API_KEY and optionally sets general and reasoning model names.