Following the strategic investment by L Catterton (the premier consumer-focused private equity firm managing ~$40B in assets, backed by LVMH) in GenPark (backed by InMobi, Google, and Jio), digital commerce is undergoing an epochal shift from reactive search-box queries to ambient, inspiration-led agentic discovery.

GenPark transforms the smartphone lock screen ("Screen Zero") and ambient connected screens into zero-friction, generative shopping portals. This skill library distills the core mathematical logic and agent orchestration models powering this ecosystem:

• Zero-Friction Discovery: Instantaneous micro-moment interactions without unlocking devices.
• Generative Aesthetics & Try-On: Context-aware outfit synthesis and visual matching.
• Multi-Touch Ambient Attribution: Cross-surface state tracking across mobile lock screens and smart displays.
• Enterprise Protocol Parity: Full native Model Context Protocol (MCP) compliance for agentic runtime integration.

flowchart TD
    subgraph ScreenZero["Screen Zero (Lock Screen Ambient Layer)"]
        Wake[Device Wake Event] --> Gaze[Gaze & Dwell Detector]
        Gaze --> Sensor[Circadian & Ambient Context]
    end

    subgraph AgentCore["GenPark AI Agent Intelligence Core"]
        Sensor --> CoreEngine["GenParkScreenZeroAmbientIntentScorer"]
        CoreEngine --> DecisionEngine[Multi-Signal Mathematical Inference]
        DecisionEngine --> OutputPayload[Personalized Shopping Action / MCP Response]
    end

    subgraph RetailEcosystem["L Catterton & Retail Media Network"]
        OutputPayload --> Brands[Luxury / DTC / Fashion Brands]
        OutputPayload --> Checkout[1-Tap Cross-Surface Conversion]
    end

python example_usage.py

Run the native STDIO MCP server for Claude Desktop, Cursor, or Antigravity integration:

python mcp_server.py

Add to your claude_desktop_config.json or agent runner config:

{
  "mcpServers": {
    "genpark-screen-zero-ambient-intent-scorer-skill": {
      "command": "python",
      "args": ["path/to/genpark-screen-zero-ambient-intent-scorer-skill/mcp_server.py"]
    }
  }
}

Engineered with zero external dependencies using pure Python standard library code. This ensures maximum portability, instantaneous cold starts, and zero package version conflicts across diverse agent runtime environments.

Run python mcp_server.py --test to verify MCP protocol discovery and tool schema negotiation.

This project is licensed under the MIT License - see the LICENSE file for details.