If you ask ChatGPT or Claude whether a 1.25-carat oval diamond quoted at $5,500 is actually a good deal, it will usually give you a polite, generic answer about the 4 Cs. It will remind you that color and clarity matter, but it won’t be able to tell you if you are getting a fair quote on the counter right now.

The reason is simple: language models understand gemology concepts, but their training data is static. They do not have live access to current wholesale price levels.

We built the OpenFacet MCP server to bridge that gap. By connecting OpenFacet’s pricing engine to AI assistants via the Model Context Protocol (MCP), your AI tools can look up exact, real-time natural diamond benchmarks before answering.

What is MCP in Plain English?

Think of a standard AI as a knowledgeable advisor without internet access. It understands how diamond pricing works in theory, but it cannot check today’s market boards.

MCP (Model Context Protocol) is an open communication standard created by Anthropic. It functions like a direct link between the AI and live data sources.

When you ask a question like:

I’m looking at a 1.23ct G VS2 emerald cut with a 1.55 length-to-width ratio for $5,200. Is that fair?

Instead of estimating or making up a figure, the AI makes a fast request to https://mcp.openfacet.net/, retrieves the interpolated market price, checks the ratio adjustment, and gives you a concrete, data-backed answer.

What the OpenFacet MCP Server Does

The server is lightweight, dependency-free, and provides three main tools:

1. get_diamond_price: Calculates per-carat and total prices for GIA-graded natural diamonds (0.30ct to 6.00ct, D–M color, FL–SI2 clarity). It avoids crude weight brackets by using log-space interpolation for exact weights, and factors in length-to-width ratio curves for fancy cuts (oval, emerald, cushion, radiant, pear, marquise, and heart).
2. get_dcx_index: Retrieves the Diamond Composite Index (DCX) and its 24-hour, 7-day, and 30-day trends.
3. get_market_depth: Looks at actual wholesale supply across color and clarity combinations so you can see how liquid or rare a stone’s profile is.

Why It’s Useful

• For Buyers: Sanity-check any in-store or online quote in seconds without digging through confusing trade lists.
• For Developers & Traders: Build automated valuation bots or inventory monitors on top of a free, unauthenticated data endpoint.
• Continuous Weight Precision: A 1.18ct stone is not the same as a 1.45ct stone. OpenFacet calculates prices continuously across every point of carat weight rather than grouping them into coarse boxes.

Try It Out

• Live Remote Endpoint: https://mcp.openfacet.net/
• Source Code: github.com/openfacet/mcp-server
• API & Developer Documentation: OpenFacet API Documentation
• ChatGPT Demo: OpenFacet ChatGPT Plugin Integration

You can plug it into Claude Desktop, Cursor, ChatGPT, or your own backend scripts in just a couple of minutes.