Get the FREE Ultimate OpenClaw Setup Guide →

taskaza

AI-powered task agent enabling natural language task management via MCP-based agents, with secure FastAPI APIs, hierarchical workflows, and a decoupled, production-ready web architecture

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio kayvanshah1-taskaza docker run -i kayvanshah1/taskaza-backend:latest \
  --env LOG_LEVEL="info" \
  --env JWT_SECRET="your-jwt-secret" \
  --env DATABASE_URL="postgresql://user:password@host:5432/taskaza" \
  --env OPENROUTER_API_KEY="your-openrouter-api-key"

How to use

Taskaza is an AI-powered task management backend with a FastAPI API surface and an accompanying React frontend. This MCP server package exposes the Taskaza backend through a Docker image that runs the full backend stack, including authentication (JWT + API keys), task CRUD operations, and an AI-driven AskBar endpoint powered by a FastMCP/OpenRouter integration. Once running, you can interact with the REST API to create, list, update, and delete tasks, manage API keys, and trigger AI-assisted actions via the /ask-agent endpoint. The system maintains conversational history for AI interactions and provides reasoning traces for transparency. To get started, run the supplied container (or use your chosen deployment method), connect to the API docs, and begin issuing requests to the task management endpoints or the AI agent endpoint to perform natural language queries like “show me my overdue tasks” or “remind me to follow up with the supplier.”

How to install

Prerequisites:

  • Docker and Docker Compose installed on your machine or server
  • Basic familiarity with Docker commands

Installation steps:

  1. Ensure Docker is running on your machine.

  2. Pull the Taskaza backend image (or use the provided Docker run command in mcp_config):

    docker pull kayvanshah1/taskaza-backend:latest

  3. Run the container with required environment variables. Replace placeholders with real values:

    docker run -i
    -e JWT_SECRET=your-jwt-secret
    -e DATABASE_URL=postgresql://user:password@host:5432/taskaza
    -e OPENROUTER_API_KEY=your-openrouter-api-key
    -e LOG_LEVEL=info
    kayvanshah1/taskaza-backend:latest

  4. Confirm the backend is reachable (e.g., via port mapping if you expose ports) and consult the API docs for available endpoints.

  5. (Optional) Configure a reverse proxy or orchestrator (Docker Compose, Kubernetes) for production deployment and proper health checks.

Additional notes

Tips and tips:

  • Replace placeholder environment variables with secure values in production (especially JWT_SECRET and DATABASE_URL).
  • If you use Docker, consider mapping ports and setting persistent storage for the database in a real deployment.
  • Ensure network accessibility between the backend and the PostgreSQL database. Use a dedicated database user with restricted permissions.
  • The /ask-agent endpoint leverages FastMCP and OpenRouter LLMs; ensure you have a valid API key for OpenRouter if required by your deployment.
  • Check the GitHub Actions CI/CD workflows referenced in the README for deployment cues and environment example configurations.
  • If you encounter CORS or authentication issues, review the API keys per user and verify that the JWT setup aligns with the client frontend expectations.

Related MCP Servers

Sponsor this space

Reach thousands of developers