todo-management

Manage multi-step tasks in canonical TODO.md with atomic writes and locking.

8|1|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/bordenet/superpowers-plus --skill todo-management-bordenet
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: todo-management
Source: https://github.com/bordenet/superpowers-plus/tree/main/skills/productivity/todo-management
Command: npx skills add https://github.com/bordenet/superpowers-plus --skill todo-management-bordenet

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Centralizes capture and persistent tracking of multi-step tasks while preventing data loss and conflicting writes by multiple agents. It prevents agents from bypassing safeguards, enforces structure for plan-level work, and makes progress and history reliably queryable.

Core Features & Use Cases

  • Persistent TODO.md storage with advisory locking, backups, atomic writes, and immutability defenses to prevent accidental or malicious overwrites.
  • Multi-step plan tracking and mirroring that creates plan-scoped task groups (#plan-<id>), mirrors long-lived plans into session tools, and enforces context-aware fields for handoff.
  • Multi-agent coordination via claim/unclaim/reap semantics, TTLs, and maintenance tooling so agents avoid duplicate work and stale locks.
  • Maintenance and recovery tools to archive, detect annihilation, run housekeeping, and surface health checks.

Quick Start

Add a P1 task: Fix production login bug and persist it to the canonical TODO.md with tags #engineering-backend #plan-auth-fix.

Frequently Asked Questions about todo-management

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I persist multi-step tasks for collaborative agents without losing data?

Persistent task tracking for collaborative agents uses a canonical TODO.md file with atomic writes, timestamped backups, and advisory locking to prevent data loss and conflicting overwrites. This ensures multi-step plans are safely stored and reliably queryable.

How does advisory locking prevent duplicate work in multi-agent task management?

Advisory locking in multi-agent task management uses claim, unclaim, and reap semantics with TTLs. This coordination mechanism prevents duplicate work by ensuring only one agent holds a lock on a task, while maintenance tooling clears stale locks.

What is the best way to track long-lived plans across multiple agents and sessions?

Tracking long-lived plans involves creating plan-scoped task groups with context-aware fields and mirroring them into session tools. This enforces structured handoff fields, making plan progress and history reliably queryable across different agents and sessions.

Can I recover overwritten tasks from a persistent TODO.md file?

Recovery for an overwritten TODO.md file relies on timestamped backups and file immutability defenses. Maintenance and recovery tools archive tasks, detect annihilation, run housekeeping, and surface health checks to restore lost data.

Does multi-agent task management work without external dependencies?

Multi-agent task management operates with zero external dependencies by integrating internal session tools and maintenance scripts. It enforces task ID allocation and structured sections natively, requiring no additional packages to manage persistent workflows.

When should I not use a canonical TODO.md for task prioritization?

A canonical TODO.md for task prioritization should be avoided when workflows require high-concurrency transactional databases rather than file-based locking. It is optimized for capturing and tracking multi-step plans, not heavy relational data processing.