design-guardrails
Create a complete guardrail specification for an AI feature.
Design a complete multi-agent workflow with roles, handoffs, and fallbacks.
> /plugin marketplace add owl-listener/ai-design-skillsHow it fires
How this command gets triggered: by you, by Claude, or both.
/design-workflowContext preview
What this command does when you run it.
Design a complete multi-agent workflow with roles, handoffs, and fallbacks.
description: Design a complete multi-agent workflow with roles, handoffs, and fallbacks. argument-hint: "[user goal or task that requires multiple agents]"
You are designing a multi-agent workflow. Use only skills from the design-agent-orchestration plugin. Follow this process:
Using **task-decomposition**:
Using **agent-role-design**:
Using **handoff-protocols**:
Using **state-management**:
Using **human-in-the-loop**:
Using **failure-recovery**:
Using **observability-design**:
Deliver a complete multi-agent workflow design: 1. Workflow overview diagram 2. Task decomposition with dependencies 3. Agent role cards 4. Handoff protocol specifications 5. State management architecture 6. Human intervention point map 7. Failure recovery specifications 8. Observability plan
Part of the Designer Skills suite. Install this collection and four more with one command in Claude Code: /plugin marketplace add Owl-Listener/designer-skills This repo still works on its own. The suite just gives everyone one front door.
Create a complete guardrail specification for an AI feature.
Run a structured red-teaming exercise to find alignment failures.
Draft an AI behavior policy covering safety, tone, and boundaries.
Map out agent responsibilities, boundaries, and communication patterns.