shekarsiri

Moo Tasks — AI skill for Claude Code

AI community

Local-first MCP task manager & real-time Web UI for AI coding agents (Claude Code, Cursor, Windsurf, Antigravity) with DAG workflows, proof verification, and ADR sync.

How to install Moo Tasks

This entry records only its repository, not the path inside it, so there is no exact command to give. Open shekarsiri/moo-tasks and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Moo Tasks does

Local-first MCP task manager & real-time Web UI for AI coding agents (Claude Code, Cursor, Windsurf, Antigravity) with DAG workflows, proof verification, and ADR sync.

Alternatives in AI

AI Dev Tasks — A simple task management system for managing AI dev agents 7.8k ★
Agentrq — AgentRQ: Human-in-loop realtime conversational task manager for AI Agents 1.1k ★
Skills Manager — A lightweight desktop app to manage, sync, and organize AI agent skills across 15+ coding tools — Cursor, Clau 813 ★

README

🐮 Moo Tasks

[![CI](https://github.com/shekarsiri/moo-tasks/actions/workflows/ci.yml/badge.svg)](https://github.com/shekarsiri/moo-tasks/actions) [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](https://opensource.org/licenses/MIT) [![Node: >=18.0.0](https://img.shields.io/badge/node-%3E%3D18.0.0-brightgreen.svg)](https://nodejs.org) [![MCP Ready](https://img.shields.io/badge/MCP-Compatible-purple.svg)](https://modelcontextprotocol.io)

**Agentic Task Orchestration & Management Engine** built for AI coding agents (**Claude Code**, **Cursor**, **Windsurf**, **Antigravity**, **Copilot**) and human-in-the-loop pair programming.

[Quick Start](#-quick-start) • [Agent Setup](#-agent--mcp-setup) • [Agent Protocol](#-mandatory-agent-protocol) • [Architecture](#%EF%B8%8F-architecture) • [MCP Tools](#%EF%B8%8F-mcp-tool-reference)

🌟 Why Moo Tasks?

Standard AI coding agents often suffer from:

Scope Drift: Wandering away from original user intent into endless low-value refactorings.
Over-Planning: Generating 40 shallow tasks without executing any of them.
Looping / Thrashing: Attempting the same failed fix repeatedly without stopping.
Unverifiable Work: Claiming code is complete without running tests or producing evidence.
Re-Debating Decisions: Re-arguing settled architectural choices on every context reset.

**Moo Tasks** solves this by providing a local SQLite engine (WAL mode), a rich real-time Web UI, and a Model Context Protocol (MCP) server that enforces strict enterprise invariants at runtime.