Ntriq Vision Product Analyzer Mcp

Analyze product images: identify items, extract specs, compare features, generate descriptions.

Vision Product Analyzer MCP

Analyze product images to identify items, extract technical specifications, compare features across variants, and generate optimized product descriptions for e-commerce. Powered by local Qwen2.5-VL — no external API required.

Parameters

Parameter	Type	Required	Description
images	array	✅	Product image URLs or base64 (up to 10 per product)
tasks	array	❌	identify, specs, compare, description (default: all)
category_hint	string	❌	Product category hint for better accuracy
competitor_images	array	❌	Competitor product images for comparison

Example Response

json

{
  "product_identified": "Mechanical Gaming Keyboard",
  "category": "Computer Peripherals",
  "specifications": {
    "switch_type": "Cherry MX Red (visible through keycaps)",
    "form_factor": "TKL (Tenkeyless)",
    "backlight": "RGB per-key",
    "estimated_dimensions": "360mm x 130mm",
    "materials": ["aluminum top plate", "PBT keycaps"]
  },
  "description": "Compact tenkeyless mechanical keyboard with Cherry MX Red switches and per-key RGB backlighting. Aluminum top plate construction. Ideal for gaming and space-constrained setups.",
  "seo_tags": ["mechanical keyboard", "tkl keyboard", "cherry mx red", "rgb keyboard", "gaming keyboard"]
}

Use Cases

• E-commerce catalog automation from manufacturer photos
• Product returns damage assessment
• Competitive product feature comparison research

Access

bash

# x402 endpoint — pay $0.05 USDC per call (Base mainnet)
POST https://x402.ntriq.co.kr/vision-product

# Service catalog
curl https://x402.ntriq.co.kr/services

x402 micropayments — USDC on Base, gasless EIP-3009