Stats
Actions
Tags
Designs a complete human-AI conversation flow for a feature, producing a design document with flow diagram, initiative map, modality specs, repair protocol, disclosure plan, and 3 transcripts.
How this command is triggered — by the user, by Claude, or both
Slash command
/model-interaction-design:design-conversation [feature or task description]The summary Claude sees in its command listing — used to decide when to auto-load this command
You are designing a human-AI conversation flow. Use only skills from the model-interaction-design plugin. Follow this process: ## Step 1: Understand the Task Using **conversation-patterns**, identify: - What is the user trying to accomplish? - What information does the AI need from the user? - What information does the user need from the AI? - What's the expected conversation length (2 turns or 20)? ## Step 2: Choose the Dialogue Structure Using **conversation-patterns**, select the appropriate structure: - Interview, co-creation, instruction-execution, exploration, or guided workflow - Jus...
You are designing a human-AI conversation flow. Use only skills from the model-interaction-design plugin. Follow this process:
Using conversation-patterns, identify:
Using conversation-patterns, select the appropriate structure:
Using mixed-initiative-flow, define:
Using multimodal-orchestration, specify:
Using conversation-patterns (repair sequences) and feedback-loops:
Using progressive-disclosure:
Deliver a complete conversation design document:
npx claudepluginhub owl-listener/ai-design-skills --plugin model-interaction-design/design-ai-featureDesign an AI-powered feature end-to-end — model selection, prompt architecture, eval framework, failure modes, cost modeling, and improvement flywheel
/design-promptGenerates a complete system prompt package for an AI feature or product, including structured prompt, constraints, examples, rationale, context spec, and test cases.
/design-chatbotGenerates a complete chatbot UI spec covering persona, dialog flows, message components, error states, and accessibility. Accepts assistant type, platform, and persona brief as arguments.
/design-workflowDesigns a complete multi-agent workflow for a user goal or task, producing overview diagram, task decomposition, agent roles, handoffs, state management, interventions, recovery, and observability plan.
/brainstormExplores user workflows using role templates to design agent systems by invoking the rcc:brainstorming-workflows skill.
/ideateStarts a collaborative design exploration for a feature or problem, guiding users through clarifying questions, approach exploration, and design documentation.