breezing

Coordinate multi-agent task execution across Lead, Worker, and Reviewer roles.

2|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/fockus/claude-skill-build --skill breezing-fockus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: breezing
Source: https://github.com/fockus/claude-skill-build/tree/main/skills/breezing
Command: npx skills add https://github.com/fockus/claude-skill-build --skill breezing-fockus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinating multi-agent team execution often requires rigid handoffs and ad hoc coordination. Breezing provides a backward-compatible alias for harness-work that enforces distinct Lead, Worker, and Reviewer roles to orchestrate tasks in a team setting.

Core Features & Use Cases

  • Role separation: Lead delegates tasks, Workers implement, Reviewers verify; supports parallelism and auto-mode.
  • Team orchestration: Provides a team-execution workflow mirroring real projects, with planning, delegation, and post-delegation validation.
  • Codex / parallel mode: Optional Codex delegation for full automation and auto-mode follow-ups.

Quick Start

Run breezing to initialize a team-execution workflow with lead, worker, and reviewer roles.

Frequently Asked Questions about breezing

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

FAQPage Schema
What is multi-agent team execution orchestration for parallel coding tasks?

Multi-agent team execution orchestration coordinates distinct Lead, Worker, and Reviewer roles to delegate, implement, and verify long-running task sequences. It enforces clear role separation and structured handoffs across planning, delegation, and review phases.

How do I coordinate parallel workers and a reviewer agent for code review?

You coordinate parallel workers and a reviewer by initializing a team-execution workflow with distinct roles. The Lead delegates tasks, Workers implement them in parallel, and the Reviewer verifies the outputs through post-delegation validation.

Does multi-agent orchestration support auto-mode rollout for long-running task sequences?

Yes, multi-agent orchestration supports auto-mode rollout for long-running task sequences. It enables optional Codex integration for full automation and manages scalable team orchestration with configurable parameters and auto-mode follow-ups.

Can I use parallel worker execution for large-scale software engineering delegation?

Yes, you can use parallel worker execution for large-scale software engineering delegation. The orchestration supports scalable team coordination, allowing multiple Workers to implement tasks concurrently while the Lead manages the delegation sequence.

What's the best way to structure planning, delegation, and review phases in multi-agent execution?

The best way to structure these phases is through a harness-work workflow that explicitly separates Lead, Worker, and Reviewer roles. This enforces clear structure across planning, delegation, and post-delegation validation, mirroring real project workflows.

Why do my multi-agent team executions lack clear role separation and validation?

Team executions lack clear role separation without an orchestration framework enforcing distinct Lead, Worker, and Reviewer roles. Using a structured harness-work workflow provides explicit role separation and post-delegation validation to prevent ad hoc coordination.