Sunex Optics MCP Server

A public Model Context Protocol server that lets AI assistants search Sunex's lens and imager catalog in natural language.

Live endpoint: https://mcp.sunex-ai.com/mcp Landing page: sunex-ai.com Transport: Streamable HTTP (MCP spec 2025-03-26). Legacy SSE endpoint at /sse preserved for older clients.

Connect in 30 seconds

Claude

Settings → Connectors → Add custom connector → paste https://mcp.sunex-ai.com/mcp

Cursor / Continue / Zed

Add to your MCP config with transport streamable-http and the URL above.

ChatGPT

Via any MCP → OpenAPI bridge as a custom GPT Action.

Five tools

Example prompts

• "Recommend a wide-angle lens for the Sony IMX577 with F/2.0 or faster."

• "I need fisheye lenses under $100."

• "What's the diagonal of the IMX477 in mm?"

• "Find lenses for a 1920×1080 sensor with 3µm pixels, 100–180° HFOV."

Architecture

Claude / Cursor / ChatGPT  →  mcp.sunex-ai.com  →  optics-online.com/api/v1
     (MCP client)         (Cloudflare Worker)      (ASP JSON API)

Thin proxy on Cloudflare Workers (free tier) over Sunex's production catalog. Streamable HTTP transport per MCP spec 2025-03-26 (with legacy SSE preserved). No auth, read-only.

Endpoints

Self-host

git clone https://github.com/Sunex-AI/Optics-mcp
cd Optics-mcp
npm install
npx wrangler login
npx wrangler deploy

Calling a tool directly (Python)

from mcp import ClientSession
from mcp.client.streamable_http import streamablehttp_client

async with streamablehttp_client("https://mcp.sunex-ai.com/mcp") as (r, w, _):
    async with ClientSession(r, w) as session:
        await session.initialize()
        result = await session.call_tool(
            "recommend_lens_for_imager",
            {"imagerPn": "IMX577", "fNumMax": 2.0}
        )

Discovery

Public manifest: https://mcp.sunex-ai.com/.well-known/mcp.json

Contributing

Issues and PRs welcome. For requests about the backend API (pricing, additional catalog fields, new endpoints), email support@sunex.com.

License

MIT — see LICENSE.