Designs human-AI conversations with patterns for turn-taking, repair sequences, grounding, and structures like interviews, co-creation, or guided workflows.
npx claudepluginhub owl-listener/ai-design-skills --plugin model-interaction-designThis skill uses the workspace's default tool permissions.
Conversation between humans and AI follows predictable structural patterns. Designing these deliberately — rather than leaving them to model defaults — is core interaction design work.
Proactively detects and restructures user messages exhibiting anti-patterns like recursive nesting, scope creep, buried asks, assumed context, stream dumps, and imprecise descriptions.
Designs and optimizes prompts for LLM apps, mastering structure, system/user messages, few-shot examples, chain-of-thought, output parsing, chaining, evaluation, and token optimization. Use for prompt engineering tasks.
Transforms vague prompts into structured ones with roles, task decomposition, output formats, constraints, and quality checks. Useful for inconsistent AI outputs, multi-step reasoning, or safety guardrails.
Share bugs, ideas, or general feedback.
Conversation between humans and AI follows predictable structural patterns. Designing these deliberately — rather than leaving them to model defaults — is core interaction design work.
Every human-AI conversation has a rhythm. The designer decides:
Conversations break down. Repair is how they recover:
Grounding is how participants establish shared understanding:
Common structural patterns for human-AI conversation: