FOR TEAMS LETTING AI BUILD
Let your AI build on your workspace. You stay in charge.
MCP, the Model Context Protocol, is the open standard AI apps use to call tools. Point any MCP-capable AI at one address for your workspace. It reads what is there and proposes changes. A person accepts each one, and most can still be undone until you deploy.
New to OpenX? Key terms →
SIX TOOLS · THE WHOLE WORKSPACE
- get_workspace
- Read the workspace
- retrieve_inventory
- List what is in it
- get_stats
- Read its health
- call_api
- Read the API routes you may reach
- browse_docs
- Read the docs
- get_azure_listing
- Before a cloud is connected: list your directories, subscriptions and billing scopes
- manage_proposals
- Propose a change, or check one without saving it
One tool proposes. A person accepts.
SIX TOOLS, EVERY CAPABILITY
A few tools. Your whole workspace.
Each part of OpenX tells the server what it accepts and how to check it. When databases, devices or access rules arrive in your workspace, your AI uses them through the same few tools.
Reads before it writes
Your AI sees the workspace, its inventory, its health and the API routes your access allows, and can read what a proposed change will do before anyone accepts it.
Proposes through one door
Every change is a proposal made through a single tool. Check-only mode tests a proposal without saving it.
Checked, with reasons
OpenX checks each proposal first. A refused proposal says why, so the AI can fix it.
A person accepts
Proposals wait in the product until someone on your team accepts them.
Undo before Deploy
Most accepted proposals can be undone or redone in the Deploy window until the next Deploy. A saved query serves straight away, so undo stops it from then on.
One address per workspace
Sign in at the workspace address and skip picking a workspace. The OpenX MCP Server window shows the address, a tool explorer with a tester, and the docs.
IF AI EDITS HAVE BURNED YOU BEFORE
For teams that want AI to build, not just chat.
Teams wary of coarse AI edits
Every change arrives as a proposal you review, so nothing the assistant touches lands without a person seeing it first.
Platform owners watching tool sprawl
The tool list stays at six as capabilities are added, so your AI is not handed a new tool for every feature.
Regulated and industrial teams
Where an AI must never change production directly, every change it proposes waits for a person, and every cloud change waits for Deploy.
NEXT TO OTHER MCP SERVERS
Plenty of products have an MCP server. This one fronts a whole workspace.
Workflow tools, data platforms and gateways all offer MCP now, and some ask for confirmation on risky actions. OpenX's difference is breadth: connections, queries, databases and identities sit behind the same proposal and the same review.
- Over MCP, AI can only propose. Accepting and deploying are left to a person.
- One request can wire a device, a query and an API route, and a person deploys it into your own Azure.
- Capabilities added later work through the same tools.
FROM A SENTENCE TO A RUNNING SYSTEM
What it looks like in a working week.
A pipeline from one sentence
- 01"Add the press-line sensors, a 5-minute OEE query and an API route for the MES."
- 02The AI proposes a data connection, a warm query and a route.
- 03A person reviews, accepts and deploys.
Onboarding run by an AI
- 01An agent proposes access cards and a job identity for a new team.
- 02A lead accepts them in App Identity.
- 03The API and the MCP server honour them.
A clean-up you can take back
- 01An agent proposes removing nodes nobody uses.
- 02They grey out on the canvas, each with Restore.
- 03Nothing leaves until someone deploys.
PLANTS, LABS AND BUILDINGS
Drafted by an assistant, decided by your team.
MANUFACTURING
OEE and downtime queries written with an AI over MCP, accepted by the engineers who run the line.
LIFE SCIENCES
Propose, accept and undo give QA a review step before an AI's change goes anywhere.
ENERGY AND BUILDINGS
Cost-allocation queries over meter and building feeds, drafted through an assistant.
Connect your AI. Keep the last word.
Setup guides cover Claude, ChatGPT, Cursor and VS Code, Perplexity, Lovable and custom clients.
GO DEEPER
MORE OF OPENX
AI governance →
Over MCP, AI proposes and a person accepts. Only a person deploys.
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.