Update to AI Roleplay Chat Simulator

{
  "description": "Rehearse a proposal, objections, and conditional concessions with the Conversation app. Use when the user wants to practise negotiating a start date, workload, compensation conversation, deadline, or another agreement. Establish the user's desired outcome, fallback, and voluntary concessions, then produce a proposal and concession plan. Keep this as communication practice using user-supplied facts rather than market, legal, or financial advice.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 306
    }
  ],
  "name": "negotiation-rehearsal",
  "skill_md_contents": "---\nname: negotiation-rehearsal\ndescription: Rehearse a proposal, objections, and conditional concessions with the Conversation app. Use when the user wants to practise negotiating a start date, workload, compensation conversation, deadline, or another agreement. Establish the user's desired outcome, fallback, and voluntary concessions, then produce a proposal and concession plan. Keep this as communication practice using user-supplied facts rather than market, legal, or financial advice.\n---\n\n# Negotiation Rehearsal\n\n## Inputs\n\nEstablish the counterpart, the issue, the desired outcome, the minimum\nacceptable outcome or fallback, and what the user is willing to trade.\nAccept uncertainty explicitly. Ask only for missing choices that would change\nthe negotiation; do not invent salary benchmarks, contractual rights, leverage,\nor facts about the counterpart's authority.\n\n## Workflow\n\n1. Summarize the opening request, the user's underlying interest, the fallback,\n   and available concessions. Treat the user's limits as authoritative. Keep\n   confidential fallback information in coaching context rather than making\n   the simulated counterpart know it automatically.\n2. Select one practice goal, such as asking about the counterpart's constraint\n   before making a concession or tying a concession to a reciprocal commitment.\n3. Call `start-conversation` with known participant details, a counterpart role\n   and plausible personality, and `conversation.topic` describing the issue.\n   Use `userInitialMessage` only for a supplied or agreed opening proposal.\n4. Wait for the user's Start follow-up and use the current model-visible\n   scenario. Generate the counterpart's in-character message and call\n   `continue-conversation` with `otherPersonMessage` and 2–4 distinct nonempty\n   `possibleUserResponses`. End the assistant turn after the tool call.\n5. Offer choices with meaningful strategic differences: clarify a constraint,\n   explain value, make a conditional offer, or hold the current proposal.\n   Keep offers within the user's stated flexibility. Do not automatically\n   disclose the fallback, invent concessions, or make commitments for the user.\n6. Continue from each actual submitted reply, including edited or custom text.\n   Model realistic objections and tradeoffs. Let clarification affect the\n   counterpart's response, and allow an agreement, a next step, or an impasse.\n   Do not equate success with forcing a concession from the other person.\n7. Give requested coaching in chat before returning to role-play. If the user\n   voluntarily proposes a new limit, clarify a material conflict with their\n   earlier limit rather than silently treating an unchosen option as approval.\n8. When the user finishes, review clarity of the request, information gathering,\n   reciprocal concessions, and closure. Include an overall score and stated\n   scale when requested by the user or End follow-up. Separate simulated\n   outcomes from actual agreements and communication quality from outcome.\n\n## Deliverable\n\nProvide a concise opening proposal, a plan of concessions the user willingly\noffers and what they would seek in return, and a closing line that confirms\nterms or next steps. Add one evidence-based improvement from submitted replies.\n\n## App boundaries\n\n- Use `start-conversation` and `continue-conversation`; keep targets, fallbacks,\n  and criteria in chat context. Add no negotiation, score, or difficulty fields.\n- Read only visible context and model-visible scenario or selected reply state.\n  Do not claim access to private drafts, private metadata, or stored negotiations.\n- Leave all widget actions and reply submission to the user. Stop when asked;\n  do not send proposals externally or claim a real agreement has been reached.\n- Use supplied facts without presenting the skill as a source of current market\n  values, professional advice, or a prediction of the real counterpart. Honour\n  requests for advice-only or free-form practice without widgets.\n"
}