tasks

This skill provides a systematic approach for AI agents to manage software development work using the Mighty (mt) task tracking framework. It is designed for developers and AI agents who require a disciplined, Git-native workflow for implementing bug fixes, feature requests, and code refactoring. By treating tasks as first-class citizens, it ensures that every code change is backed by a specific requirement, documented with ephemeral progress updates, and justified through formal design decisions. The skill mandates a high-signal communication style, focusing on clear intent, source tracking, and verifiable evidence of completion.

• Streamlines task creation and management using the mt CLI, linking directly to technical specifications and design documents.

• Implements a progressive disclosure workflow, encouraging short, actionable progress comments that explain the 'why' behind implementation choices.

• Facilitates explicit design decision recording, allowing agents to document alternative paths and final trade-offs, which is crucial for long-term codebase maintainability.

• Ensures accountability through evidence-based work, where each task is mapped to specific code files, commits, and test suites via the link command.

• Enforces a clean closeout protocol that validates the state of the task, missing requirements, and provides a clear resolution summary for future audits.

• Use this skill whenever you are tasked with modifying the repository, whether implementing a new feature or fixing a bug.

• Always start by querying the current task state using mt work or mt mine to ensure synchronization with ongoing development efforts.

• When creating a task, use the provided templates to ensure goals, acceptance criteria, and constraints are clearly defined.

• Utilize mt comment to leave breadcrumbs during your coding session, especially when navigating complex refactoring or trade-offs.

• Before finishing your work, utilize the mt closeout and mt commit commands to ensure the repository remains in a clean, documented, and deployable state.

• Inputs typically include task IDs or feature specifications; outputs are structured task updates, linked evidence commits, and finalized resolution summaries.