Skip to content
CerebroChain

MCP Connector

Use CerebroChain from Claude and other AI assistants.

CerebroChain ships a remote MCP (Model Context Protocol) server, so AI assistants can compare shipping rates, query your inventory and fleet, run demand forecasts, and read live supply-chain market data — in plain language, with your permission.

Setup

The server lives at https://cerebrochain.com/mcp (Streamable HTTP transport).

  1. In Claude (web or desktop), open Settings → Connectors → Add custom connector.
  2. Paste the server URL: https://cerebrochain.com/mcp.
  3. Claude opens the CerebroChain authorization page. Sign in with your CerebroChain account (the same email and password you use at app.cerebrochain.com) and click Approve.
  4. That's it — ask Claude something from the examples below. No CerebroChain account yet? Start a 14-day free trial (no credit card).

How authorization works

The connector uses OAuth 2.1 with PKCE and Dynamic Client Registration — the standard MCP authorization flow. Your password is only ever entered on cerebrochain.com; the assistant receives a scoped, revocable token, never your credentials. Tokens map to your workspace, so the assistant sees exactly what your account can see, under one scope (mcp).

Building your own agent instead? Machine-to-machine callers can alternatively send a tenant API key in the X-API-Key header. Discovery metadata is published at /.well-known/oauth-authorization-server and /.well-known/oauth-protected-resource/mcp.

What the assistant can do

17 tools, all read-only — the connector never mutates your data:

  • Shipping & logistics — compare carrier rates (UPS, FedEx, USPS, DHL), freight rates across 85+ carriers, AI route optimization.
  • Your workspace — query inventory, fleet status, vehicles, drivers, maintenance, and routes.
  • Market intelligence — port congestion (NOAA), global shipping lanes, forex, commodity prices, FRED economic indicators.
  • Cerebrix AI — demand-forecast questions and supply-chain inference in natural language.

Example prompts

Try these once connected:

  • "Compare shipping rates for a 12 lb package from 89501 to 30301." → a carrier-by-carrier rate table with the cheapest and fastest options highlighted.
  • "What's my current inventory looking like? Anything low or out of stock?" → a live read of your workspace inventory with stock-status filters applied.
  • "Which major ports are congested right now, and how should that change my inbound planning?" → current port congestion readings plus the assistant's analysis against your lanes.
  • "Optimize a delivery route for these five stops, tightest time-windows first." → an ordered route with ETAs from the route-optimization engine.

Limitations

  • Read-only. The connector cannot create, edit, or delete anything in your workspace. Write operations stay in the console.
  • Your account's scope. Results are limited to the workspace the signed-in account belongs to.
  • Rate limits apply. Bursty automated traffic gets throttled; sustained high-volume API use belongs on the developer API instead.
  • Fresh workspaces look empty. Inventory and fleet tools return real (possibly empty) data until you've onboarded records — market-data and rate tools work immediately.
  • Access tokens expire after 1 hour and refresh automatically; reconnect from the assistant's connector settings if a session is revoked.

Support

Stuck, or found a bug? Email [email protected], call 775-719-9177, or use the contact form. We reply within one business day. Security issues: see vulnerability disclosure.