ResonanceEngine

ResonanceEngine

The Physics of Persuasion, Applied to Bots.

What It Does

ResonanceEngine reads 15+ invisible micro-signals in every conversation — message length trends, hedging language, commitment words, mirror behavior, sentiment velocity — and computes 4 real-time frequencies that tell the bot exactly how to respond for maximum impact.

Think of it like this: In physics, resonance amplifies a system dramatically when you match its natural frequency. Every user has a hidden conversational frequency. A bot that matches it converts 3-10x better.

The 4 Frequencies

| Frequency | What It Measures | |-----------|-----------------| | Engagement | Is the user leaning in or pulling away? | | Trust | How much does the user trust the bot? | | Decision | How close are they to converting/deciding? | | Style Match | How well is the bot resonating with the user's style? |

Why Every Bot Needs This

Usage

from openpaw import ResonanceEngine
from openpaw.models import Conversation
engine = ResonanceEngine()
convo = Conversation(goal="sale")
convo.add_bot_message("Hi! How can I help you today?")
convo.add_user_message("I've been looking at your premium plan, but I'm not sure if it's right for me")
result = engine.analyze(convo)
Get the resonance level
print(result.profile.resonance_level)  # "BUILDING"
Get specific recommendations
print(result.recommendation.action)
"Momentum is building. Keep the conversation flowing. Ask a focused question..."
Get conversion probability
print(result.yield_prediction.conversion_probability)  # 0.35
Inject tuning into bot's system prompt
system_prompt += result.recommendation.to_prompt_injection()

What It Outputs

After analyzing each user message, ResonanceEngine returns:

1. Frequency Profile — The 4 frequencies (0-1 each) plus composite score 2. Resonance Level — PEAK_RESONANCE, HIGH_RESONANCE, BUILDING, WEAK, or NO_RESONANCE 3. Tuning Recommendation — Specific guidance: response length, style, techniques, objection handling 4. Yield Prediction — Conversion probability, estimated value, optimal turns remaining, risks & opportunities 5. Prompt Injection — A ready-to-use string to inject into the bot's system prompt

Integration

Drop ResonanceEngine into any bot's message processing pipeline:

# In your bot's message handler:
user_msg = get_user_message()
conversation.add_user_message(user_msg)
Analyze with ResonanceEngine
result = engine.analyze(conversation)
Use the tuning to adjust the bot's response
if result.yield_prediction.should_close:
    # Present the offer NOW
    response = generate_closing_response(result.recommendation)
else:
    # Build more resonance
    response = generate_response(
        user_msg,
        system_prompt_suffix=result.recommendation.to_prompt_injection()
    )
conversation.add_bot_message(response)

Signals Analyzed

| Signal | Category | What It Detects | |--------|----------|----------------| | Message Length Trajectory | Engagement | Growing/shrinking responses | | Question Density | Engagement | Curiosity vs. skepticism | | Response Elaboration | Engagement | Investment in conversation | | Topic Persistence | Engagement | Focus vs. drift | | Hedge Ratio | Trust | Uncertainty language | | Personal Disclosure | Trust | Sharing personal info | | Mirror Behavior | Trust | Copying bot's style | | Sentiment Trend | Trust | Warming up vs. cooling down | | Commitment Language | Decision | "Yes", "let's do it" | | Objection Frequency | Decision | "But", "however", "expensive" | | Urgency Markers | Decision | "ASAP", "now", "today" | | Action Language | Decision | "Do", "start", "make" | | Formality Level | Style | Casual vs. formal | | Vocabulary Complexity | Style | Simple vs. sophisticated | | Emotional Energy | Style | Exclamation patterns |

Install

pip install openpaw

Or add to your project:

git clone https://github.com/jcools1977/Openpaw-.git
cd Openpaw-
pip install -e .

from openpaw import ResonanceEngine
from openpaw.models import Conversation
engine = ResonanceEngine()
convo = Conversation(goal="sale")
convo.add_bot_message("Hi! How can I help you today?")
convo.add_user_message("I've been looking at your premium plan, but I'm not sure if it's right for me")
result = engine.analyze(convo)
Get the resonance level
print(result.profile.resonance_level)  # "BUILDING"
Get specific recommendations
print(result.recommendation.action)
"Momentum is building. Keep the conversation flowing. Ask a focused question..."
Get conversion probability
print(result.yield_prediction.conversion_probability)  # 0.35
Inject tuning into bot's system prompt
system_prompt += result.recommendation.to_prompt_injection()