mad-tasks

What problem does it solve?

This Skill eliminates the tedious, error-prone manual work of breaking down feature design artifacts (specifications, technical plans, data models) into structured, dependency-ordered implementation task lists, ensuring no requirements are missed and work can be safely parallelized across team members.

Core Features & Use Cases

• Adaptive Template Selection: Auto-detects the appropriate task template tier (minimal/standard/full) based on feature complexity to optimize token usage, with manual override options for team preferences.
• End-to-End Quality Gates: Includes built-in requirement coverage verification and 3-lane review (completeness, dependency correctness, coverage) to catch missing tasks, circular dependencies, or orphaned requirements before implementation begins.
• Parallelization Support: Generates dependency graphs, identifies parallel execution opportunities, and annotates tasks with parallel groups and file ownership to enable concurrent work without conflicts.
• Use Case: For a new payment processing feature with 4 user stories, the skill automatically generates a prioritized tasks.md with setup, foundational, per-story implementation, and test tasks, plus clear parallelization hints for the engineering team.

Quick Start

Invoke the mad-tasks skill with the current feature's design artifacts to generate a ready-to-execute, dependency-ordered tasks.md file.