agentleFS
mtarcuremtarcure/claude-vibe-squad/.agents/skills/conversation-design/SKILL.md

conversation-design

Skill220 starsChanged 19 days agoMITView on GitHub

Use when specifying a voice or chat agent's dialogue before implementation: map intents and turns, write sample flows, define remembered context, cover no-input, no-match, ambiguity, correction, human-handoff recovery, and require explicit confirmation for consequential actions.

The file

Files
SKILL.md

Conversation Design

Design a conversational flow — intents, turns, prompts, and recovery — so a voice/chat agent handles real dialogue, not just the happy path.

Steps

  1. Map the user intents and the jobs each conversation must complete; write the sample dialogues first.
  2. Design turn-taking: what the agent asks, confirms, and remembers across the flow; keep prompts short and answerable.
  3. Design error and recovery paths: no-input, no-match, ambiguity, correction, and escape-to-human.
  4. Define confirmation strategy for consequential actions (explicit confirm before anything irreversible or costly).
  5. Specify persona and tone as constraints, and mark any regulated/consequential flow (payments, health, identity) for review.

Acceptance

  • Happy path and recovery paths (no-input/no-match/ambiguity/handoff) are all designed.
  • Consequential actions require explicit confirmation.
  • Persona/tone are defined; regulated flows are flagged for review.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.