Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Orchestrates 5-agent Claude Code teams (1 Lead + 4 Teammates) across any domain using four canonical pipeline patterns. Each teammate is assigned a domain-specific role card, and the Lead coordinates handoffs, validates outputs, and synthesizes the final deliverable. Works for software, content, research, data analysis, and business strategy out of the box.
Structured role cards and a strict 6-step coordination protocol prevent the context leakage and uncoordinated output that occur when prompting multiple agents ad hoc.
When to use it
- Running a parallel research review with four specialized agent perspectives
- Building a feature end-to-end from architecture through code review
- Iterating a content draft between a writer and editor agent until it meets criteria
- Processing a large codebase audit by fanning out tasks across four agents simultaneously
- Drafting a business strategy with independent financial modeling and risk assessment tracks
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Build and review a new authentication module using the software-dev preset
- 1Lead confirms objective, domain preset, pipeline pattern (sequential), constraints, and definition of done with user
- 2Lead builds Workflow Instance Spec and creates team: architect, developer, tester, reviewer
- 3Lead spawns architect teammate with task to produce a design document for the auth module
- 4Lead relays design doc to developer, who produces source code; then relays code to tester for test suite
- 5Lead relays test results to reviewer for code review report, then synthesizes all artifacts into final deliverable
Design document, source code, test suite, and code review report delivered with traceability to each definition-of-done criterion
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.