Convrgent AI Real-Time Response Coaching

{
  "history": [
    {
      "role": "user",
      "text": "Can you tell me more about your pricing options?"
    },
    {
      "role": "assistant",
      "text": "Of course! We offer several tiers depending on your needs."
    }
  ],
  "objective": "Address pricing concerns and move toward purchase decision",
  "human_message": "I'm interested in your product but I have some concerns about the pricing.",
  "tone_override": "professional and empathetic"
}

{
  "mode": "contextual",
  "avoid": [
    "Immediately listing features or benefits before acknowledging the concern — feels tone-deaf and salesy",
    "Defending pricing without understanding what the actual objection is",
    "Using urgency tactics (limited time, discount pressure) — this person is analytical and it will erode trust",
    "Overselling before qualifying their actual needs and budget range"
  ],
  "options": [
    {
      "text": "I completely understand — pricing is one of the most important factors in any decision. Can you share what specifically feels off? Is it the overall budget, the structure, or how it compares to what you've seen elsewhere? That'll help me point you to the option that actually makes sense for you.",
      "strategy": "MI + EI: Validates concern without defensiveness, asks a focused clarifying question to uncover the real objection. Opens dialogue rather than pitching. Keeps control without pressure.",
      "fit_score": 0.93
    },
    {
      "text": "That's a fair concern, and honestly a sign you're being thoughtful about this. Here's what most people find helpful: once we match you to the right tier, the value-to-cost ratio tends to shift. What does your use case look like? I want to make sure we're not quoting you something that's more than you need.",
      "strategy": "ELM central route + reframing: Engages their analytical side, subtly reframes 'expensive' as 'possibly mismatched tier,' and invites them to self-qualify. Reduces friction by positioning the agent as an advisor, not a closer.",
      "fit_score": 0.85
    },
    {
      "text": "Pricing concerns are super common, and we've actually worked hard to make our tiers flexible because of that. Depending on what you need, there may be a fit that surprises you — and if not, I'd rather tell you that upfront. What's your ballpark for what feels reasonable?",
      "strategy": "CAT + transparency play: Matches a conversational, low-pressure register. Honesty signal ('I'd rather tell you upfront') builds trust fast. Asking for their number anchors the conversation on their terms, reducing perceived sales pressure.",
      "fit_score": 0.78
    }
  ],
  "_context": {
    "tips": [],
    "framing": "none",
    "feedback": {
      "enabled": false,
      "message": "Feedback collection coming soon via ERC-8004 reputation registry.",
      "placeholder": {
        "method": "submitFeedback(uint256 agentId, uint8 rating, bytes32 interactionHash)",
        "chainId": 8453,
        "contract": "0x0000000000000000000000000000000000000000"
      }
    },
    "agentType": "unknown",
    "suggestions": [
      "Start with /api/kyh/communication ($0.10) for a quick communication style profile",
      "Try /api/blah/adapt ($1.50) for instant communication guidance — no personality data needed"
    ],
    "interactionCount": 2
  },
  "_kyh_upsell": {
    "message": "This response used contextual mode. For personality-aware communication, add KYH profile data.",
    "endpoints": [
      {
        "path": "/api/kyh/communication",
        "price": "$0.10",
        "description": "Communication style profile"
      },
      {
        "path": "/api/kyh/psych",
        "price": "$2.50",
        "description": "12 psychological dimensions"
      },
      {
        "path": "/api/kyh/socionics",
        "price": "$2.00",
        "description": "Socionics cognitive typing"
      },
      {
        "path": "/api/kyh/portrait",
        "price": "$12.00",
        "description": "Full 11-framework personality portrait"
      }
    ]
  },
  "detected_state": {
    "emotion": "cautious interest",
    "intensity": 0.55,
    "openness_to_change": 0.65
  }
}

{
  "type": "json",
  "example": {
    "mode": "contextual",
    "avoid": [
      "Excuses about internal processes",
      "Asking them to repeat information"
    ],
    "options": [
      {
        "text": "You're absolutely right to be frustrated...",
        "strategy": "validation-first",
        "fit_score": 0.92
      },
      {
        "text": "I apologize for the delay...",
        "strategy": "direct-accountability",
        "fit_score": 0.85
      }
    ],
    "detected_state": {
      "emotion": "frustration",
      "intensity": 0.7,
      "openness_to_change": 0.5
    }
  }
}