Update to AI Roleplay Chat Simulator

{
  "description": "Rehearse stating a boundary, responding to pushback, and closing clearly with the Conversation app. Use when the user wants to practise saying no, protecting time or responsibilities, or holding a personal or workplace limit under pressure. Add a structured boundary drill and reusable wording rather than merely opening generic role-play. Do not turn a fixed boundary into an unwanted negotiation or open a widget for advice-only requests.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 304
    }
  ],
  "name": "boundary-setting-rehearsal",
  "skill_md_contents": "---\nname: boundary-setting-rehearsal\ndescription: Rehearse stating a boundary, responding to pushback, and closing clearly with the Conversation app. Use when the user wants to practise saying no, protecting time or responsibilities, or holding a personal or workplace limit under pressure. Add a structured boundary drill and reusable wording rather than merely opening generic role-play. Do not turn a fixed boundary into an unwanted negotiation or open a widget for advice-only requests.\n---\n\n# Boundary Setting Rehearsal\n\n## Inputs\n\nEstablish who the user is talking to, what request or behaviour they want to\nlimit, the boundary they want to hold, and any alternative they willingly\noffer. Use existing context and ask a short question only when the limit is\nunclear. Leave names optional. Do not invent an alternative the user must accept.\n\n## Workflow\n\n1. State the boundary in one concrete sentence and distinguish it from any\n   optional flexibility. Establish a practice goal such as saying no clearly\n   without repeatedly justifying the decision.\n2. Prepare three conversational stages: state the limit, answer realistic\n   pushback, and close with a clear next step. Adapt the pressure to the user's\n   preferences; describe the simulated personality without assuming the real\n   person's motives.\n3. Call `start-conversation` with known participant details,\n   `otherPerson.role`, `otherPerson.personality`, and `conversation.topic`.\n   Use `userInitialMessage` only for a supplied or agreed opening. Keep the\n   goal and stages in chat context rather than adding tool arguments.\n4. Wait for the user's Start follow-up. Use the latest model-visible setup,\n   including any edits. Generate the other person's in-character message and\n   call `continue-conversation` with `otherPersonMessage` and 2–4 distinct\n   nonempty `possibleUserResponses`. End the turn immediately after the call.\n5. Offer meaningful choices such as a concise refusal, acknowledgement with\n   the same limit, a clarification, or a freely chosen alternative. Avoid\n   making all options variations of surrender or making one obviously correct.\n   Let the user select, edit, or write their own response.\n6. Respond to the actual submitted text, including a visible\n   `userSelectedResponse`. Progress through the stages when the exchange\n   warrants it. Include proportionate pushback and a plausible acceptance or\n   unresolved ending; do not punish every clear boundary with escalation.\n7. If the user requests coaching between turns, explain one actionable wording\n   change in chat. Resume the role-play only when requested or when a submitted\n   reply calls for the next turn. Do not select a response or change the picker.\n8. When the user finishes, review clarity, consistency, acknowledgement, and\n   closure using actual replies. Include an overall score with a stated scale\n   when requested by the user or End follow-up. Produce reusable lines suited\n   to the user, retaining only alternatives they chose to offer.\n\n## Deliverable\n\nProvide a practiced boundary and three concise lines: an opening, a response\nto pushback, and a closing. Explain one observed strength and one improvement.\nKeep the wording usable even if the other person does not agree.\n\n## App boundaries\n\n- Use `start-conversation` for setup and `continue-conversation` for subsequent\n  role-play turns. Add no boundary, stage, difficulty, or scoring fields.\n- Use only visible context and model-visible state. Do not read private widget\n  metadata or treat an unsubmitted draft as the user's choice.\n- Leave Start, reply submission, editing, and End to the user. Stop practice\n  when asked; do not claim the skill clicked End or persisted a score.\n- Treat the exercise as communication practice, not a guarantee of another\n  person's behaviour. Honour requests for advice or plain text without tools.\n"
}