# genpark-agentic-multimodal-product-return-fraud-detector-skill

[![Official Hub](https://img.shields.io/badge/GenPark-Official%20Hub-6366f1?style=for-the-badge&logo=compass&logoColor=white)](https://www.genpark.ai)
[![LinkedIn](https://img.shields.io/badge/LinkedIn-GenPark-0A66C2?style=for-the-badge&logo=linkedin&logoColor=white)](https://www.linkedin.com/company/92955550)
[![Zero External Dependencies](https://img.shields.io/badge/Dependencies-Zero%20(Pure%20Stdlib)-success?style=for-the-badge)](https://www.genpark.ai)
[![MCP Compliant](https://img.shields.io/badge/MCP-JSON--RPC%202.0-blue?style=for-the-badge)](https://www.genpark.ai)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow?style=for-the-badge)](LICENSE)

> **Official Website**: [https://www.genpark.ai](https://www.genpark.ai)  
> **LinkedIn**: [https://www.linkedin.com/company/92955550](https://www.linkedin.com/company/92955550) ([https://www.llinkedin.com/company/92955550](https://www.llinkedin.com/company/92955550))

AI Multimodal E-Commerce Return Fraud & Wardrobing Abuse Detector. Correlates return velocity, carrier scan discrepancies, product tag tampering indicators, and customer lifetime return ratios to compute a risk index and enforce restocking fees or dispute escalation.

---

## 🌐 Official GenPark AI Ecosystem Links

- **GenPark Official Website**: [https://www.genpark.ai](https://www.genpark.ai)
- **LinkedIn Official Page**: [https://www.linkedin.com/company/92955550](https://www.linkedin.com/company/92955550) ([https://www.llinkedin.com/company/92955550](https://www.llinkedin.com/company/92955550))

---

## 🌟 Key Features

- **100% Zero External Dependencies**: Runs entirely on the Python 3.9+ standard library.
- **Model Context Protocol (MCP) Standard**: Native support for JSON-RPC 2.0 `initialize`, `tools/list`, and `tools/call`.
- **Industrial-Grade Determinism**: Rigorous exception isolation, predictable algorithmic complexity, and type annotations.
- **Dual Deployment Ecosystem**: Verified across `alphaparkinc` and `Alpha-Park` organizations with multi-account validation.

---

## 🚀 Quick Start

### 1. Direct Python SDK Usage

```python
from client import AgenticProductReturnFraudDetector

def main():
    detector = AgenticProductReturnFraudDetector()
    claim = {
        "order_id": "ORD-4481",
        "claimed_reason": "did not like color",
        "item_price_usd": 480.0,
        "carrier_weight_kg": 0.3,
        "expected_weight_kg": 1.5,
        "days_held": 29
    }
    history = {"total_orders": 3, "past_returns_count": 2, "past_fraud_disputes": 0}
    res = detector.evaluate_return_request(claim, history)
    print("--- Return Fraud Sentinel Evaluation ---")
    print(f"Order: {res['order_id']} | Risk Score: {res['composite_fraud_risk_score']}/100")
    print(f"Action: {res['enforced_action']}")
    print(f"Restocking Fee: {res['restocking_fee_percentage']}%")

if __name__ == "__main__":
    main()

```

### 2. Run as Model Context Protocol (MCP) Server

Start standard JSON-RPC 2.0 server over `stdio`:

```bash
python mcp_server.py
```

Execute embedded test harness:

```bash
python mcp_server.py --test
```

---

## 🛠️ MCP Tool Specification

Inspect [`skill.json`](skill.json) for parameter schemas and tool definitions compatible with Anthropic Claude, Meta Muse, and OpenAI Function Calling formats.

---

## 📜 License & Community

Licensed under the [MIT License](LICENSE). Copyright © 2026 [GenPark AI](https://www.genpark.ai).
Follow us on LinkedIn: [https://www.linkedin.com/company/92955550](https://www.linkedin.com/company/92955550) | [https://www.llinkedin.com/company/92955550](https://www.llinkedin.com/company/92955550)