agent-coordination

Orchestrate PLAN.md implementation phases with checkpoint validation and user-confirmed transitions.

Updated Dec 28, 2025
One-click install
npx skills add https://github.com/I2olanD/dotfiles --skill agent-coordination-i2oland
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-coordination
Source: https://github.com/I2olanD/dotfiles/tree/main/.config/opencode/skill/agent-coordination
Command: npx skills add https://github.com/I2olanD/dotfiles --skill agent-coordination-i2oland

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates management of plan-driven work by enforcing phased execution, checkpoint validation, and user-confirmed transitions, reducing cognitive load and ensuring traceability.

Core Features & Use Cases

  • Phase-by-phase execution of PLAN.md with checkpoint validation
  • Delegate tasks to subagents (parallel and sequential) with clear summaries
  • Track progress using todowrite and manage phase boundaries with user confirmation
  • Handle blocked states with configurable options and fail-safes

Quick Start

Activate this skill to orchestrate PLAN.md driven work. For each phase: read the current phase tasks, load them into todowrite, delegate tasks to subagents (parallel or sequential), summarize results, update phase progress, and require user confirmation to advance to the next phase.

Frequently Asked Questions about agent-coordination

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

FAQPage Schema
How do I automate plan execution with phase checkpoints?

Plan execution with phase checkpoints is automated by orchestrating phase-by-phase implementation tasks, enforcing checkpoint validation, and requiring user confirmation before advancing. This structured progress tracking reduces cognitive load and ensures traceability across projects.

What is the best way to delegate tasks to subagents during plan execution?

Task delegation to subagents during plan execution is handled by dispatching parallel and sequential tasks with clear prompts, loading tasks into todowrite, summarizing results, and managing blocked states with configurable fail-safes.

How does checkpoint validation work in phased implementation workflows?

Checkpoint validation in phased implementation workflows enforces structured progress by requiring user-confirmed transitions between phases, ensuring all current phase tasks are validated and summarized before advancing to the next phase.

Can I track plan execution progress using todowrite across multiple phases?

Tracking plan execution progress using todowrite across multiple phases is supported by loading current phase tasks, updating progress after task delegation, and managing phase boundaries with user confirmation at each transition.

Do I need a PLAN.md file to use phase management for plan execution?

A PLAN.md file is required for phase management, as the skill orchestrates PLAN.md driven workflows by reading current phase tasks, delegating them to subagents, and enforcing checkpoint validation before proceeding.

What happens when a blocked state occurs during task delegation?

When a blocked state occurs during task delegation, the skill manages it with configurable options and fail-safes, ensuring phase execution pauses and user confirmation is required before retrying or adjusting the plan.