Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
A production playbook for running multi-agent teams with defined roles, structured task states, and handoff protocols. Covers the full lifecycle from task inbox through build, review, and completion. Designed for sustained workflows where agents depend on each other's output across multiple tasks.
Structured handoff requirements and mandatory review steps prevent the compounding quality drift that occurs when agents pass work informally or skip verification.
When to use it
- Setting up a builder-reviewer agent pair for code generation and validation
- Managing task state transitions across a distributed agent team
- Defining structured handoff messages between specialized agents
- Running parallel research tasks with a reviewer aggregating results
- Establishing quality gates before shipping agent-produced artifacts
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Build and review a new authentication module using a 2-agent team
- 1Orchestrator creates a task record with ID, description, and output path
- 2Orchestrator spawns builder agent with task spec and artifact destination
- 3Builder produces auth module at /shared/artifacts/auth/ and sends handoff message with file paths, test commands, and known gaps
- 4Orchestrator spawns reviewer agent with artifact location and acceptance criteria
- 5Reviewer checks edge cases against spec and returns pass or feedback for rework
Reviewed auth module artifact with reviewer sign-off, task marked Done in lifecycle tracker