genpark-purchase-intent-scorer-skill

# genpark-purchase-intent-scorer-skill > **GenPark AI Agent Skill** -- Score visitor purchase intent from behavioral signals and recommend personalized interventions. ## Features - 10-signal scoring model: cart adds, product views, time on site, search queries, recency, referral intent, and more - Intent tiers: Hot (80+) / Warm (55-79) / Cool (30-54) / Cold (<30) - Platform-specific interventions per tier (exit-intent popups, social proof, retargeting) - Batch scoring for visitor prioritization - Full signal breakdown for analytics ## Quick Start ```python from client import PurchaseIntentClient client = PurchaseIntentClient() result = client.score( session={"product_views": 4, "cart_adds": 1, "time_on_site_sec": 300, "is_returning": True}, historical={"past_purchases": 2}, context={"referral_source": "google_shopping"}, ) print(f"Score: {result['intent_score']} | Tier: {result['intent_tier']}") print(result["recommended_interventions"]) ``` ## Installation ```bash python example_usage.py # No external dependencies ``` --- Built by [GenPark](https://genpark.ai) | [alphaparkinc](https://github.com/alphaparkinc)