FOR THE AI YOUR TEAM ALREADY USES
Bring the AI you have. Connect it in minutes.
Claude, ChatGPT, Cursor, VS Code, Perplexity, Lovable or your own client: each connects to the same OpenX MCP Server, with a setup guide in the docs.
New to OpenX? Key terms →
TWO ADDRESSES · ONE SERVER
- openx.fathym.com/ox-api/mcp
- All your workspaces; pick one at sign-in
- /ox-api/mcp/w/…
- One workspace, no picking
- Sign-in
- OAuth in the browser, or an MCP key
- Rights
- Re-checked each time the token refreshes
Every client reaches the same tools.
CLIENTS WITH A SETUP GUIDE
Works with the clients the docs cover.
Claude.ai
Add a custom connector, paste the address and sign in. On Team and Enterprise, an owner adds it first.
ChatGPT
Add a custom MCP server and sign in in the browser.
Claude Code and Claude Desktop
One command from fAI, Fathym's free command-line tool, sets it up, or add the server by hand with a key on the workspace address.
Cursor and VS Code
One fAI command, or a short JSON entry with a key. VS Code needs version 1.101 or later.
Perplexity and Lovable
Add a remote connector and sign in. In Lovable it works in chat, not inside a published app.
Your own client
JSON-RPC over Streamable HTTP with a key, or OAuth discovery for clients that support it.
IF EVERYONE ON THE TEAM USES A DIFFERENT AI
For teams that will not all pick the same AI.
Mixed teams
Engineers in Cursor, analysts in ChatGPT, leads in Claude: one workspace behind all of them.
App makers
Lovable in chat, and no-code tools such as n8n, Make, Dify, Langflow and Flowise covered together on one docs page.
Platform owners
One server to govern, whatever client a person brings.
HOWEVER THEY CONNECT
Different clients, the same rules.
Whichever client connects, it gets the same tools and the same rights as the person who signed in. Over MCP, AI can only propose, and a person accepts.
- Browser sign-in uses OAuth with PKCE. No client secret to hand around.
- Each token refresh re-checks the person's membership and rights, and never adds one.
- MCP keys last a day, a week or 30 days, and work only on one workspace's address.
FIRST CONNECTION
From a pasted address to a first proposal.
Claude, in a browser
- 01Add OpenX as a custom connector in Claude.ai.
- 02Sign in and pick a workspace.
- 03Ask what is in it.
Cursor, from the terminal
- 01Run the fAI install command the OpenX MCP Server window gives you.
- 02Cursor connects to your workspace.
- 03Ask it to draft a warm query.
A proposal, reviewed
- 01Any client proposes a change through the same tool.
- 02OpenX checks it.
- 03A person accepts it in OpenX.
WHOEVER IS ASKING
The AI each person already works in.
ENGINEERING
Cursor, VS Code or Claude Code, next to the code they are already writing.
OPERATIONS
ChatGPT or Claude, asking about the line in plain English.
APP MAKERS
Lovable and no-code tools, building on the same workspace.
Pick your client. Paste one address.
Each guide walks through setup for its client, step by step.
MORE OF OPENX
AI governance →
Over MCP, AI proposes and a person accepts. Only a person deploys.
OpenX MCP Server →
A few tools let any MCP-capable AI read your workspace and propose changes to it.
API Manager →
Your workspace is already an API. Decide who may call it and what each caller reaches.
Warm queries →
Save a question over live data and serve the answer to dashboards, apps and AI.
Provenance →
A record you can prove, kept in your own cloud, with corrections and signatures.
Your cloud →
Connect your Azure, build the landing zone, and the rest of OpenX runs inside it.
Databases →
A Cosmos DB store in your subscription, with a governed API for each container.
App Identity →
Your own sign-in directory, identities for your jobs, and access in one window.
Licenses →
Sell what you build on your own Stripe, and turn a subscription into a plan's rights.
IoT →
IoT Hub and Data Explorer in your Azure, feeding queries, APIs and AI.
Communications →
Send mail through your own relay, with the password in your vault and sends on record.
App building →
Data, sign-in, access, an API and billing: your app's backend, already there.