graph_reasoner.py

from sentence_transformers import SentenceTransformer
import numpy as np

class GraphReasoner:
    def __init__(self):
        self.model = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")

    def reason(self, text):
        """
        Basic reasoning engine — evaluates text and returns contextual insight.
        """
        keywords = ["motionboys", "empire", "dream", "memory", "future", "ai", "build"]
        text_lower = text.lower()

        if any(k in text_lower for k in keywords):
            return "That connects to your vision — expanding Motion Empire through innovation."
        elif "who" in text_lower:
            return "I’m Aventra OS, your reasoning interface and contextual brain."
        elif "goal" in text_lower:
            return "Every stored goal becomes part of your motion memory network."
        else:
            # Generate embedding-based reasoning score for simple contextual response
            vec = self.model.encode(text)
            meaning_score = np.mean(vec)
            if meaning_score > 0.01:
                return "Got it — this insight has been mapped into your reasoning graph."
            else:
                return "Noted. Your context has been added to memory."