Deploy from a sentence.
Connect Claude, Cursor, VS Code, or any MCP-capable assistant to your Lexoya account. It lists your servers, deploys your repositories, reads build logs, and manages your databases — over the Model Context Protocol, straight from your chat window.
Demo: user asks "Deploy my repo github.com/me/my-app — write a Dockerfile if it needs one." The assistant calls create_project (name "my-app", type docker, branch main, envs PORT=8080), gets a project id and queued pipeline, then calls get_pipeline_logs and reports clone, build, push, deploy succeeded — the app is live at https://a12b3c.lexoyacloud.ir.
A real exchange — your prompt, the tool calls, the live result.
One protocol between you and your cloud
MCP is the open standard that lets AI assistants call tools. Lexoya exposes 29 of them — everything the panel does, callable from a chat.
Pick an assistant
Claude Desktop, Cursor, VS Code, or any MCP-capable agent. If it speaks streamable HTTP, it works.
Add one endpoint
api.lexoya.com/v1/mcp with your API key. That's the whole setup — no plugin, no agent config.
Ask for what you need
“Deploy my repo”, “why did this fail?”, “create a Postgres for me”. The assistant picks the tools and runs the workflow.
What you can ask
Four questions people actually ask. Each one is a handful of tool calls your assistant runs against your account.
youDeploy my project and tell me when it's live.
youCreate a Postgres database for my project and connect them.
youWhy did my last deployment fail?
youHow much balance do I have left and how long will it last?
From one message to a running app
Each flow below is a handful of MCP tool calls your assistant runs for you — across the whole Lexoya platform, not just one service.
- 1
Ship an app and its database
youDeploy my repo and create a Postgres for it.
create_projectdeploy_projectcreate_quick_appget_quick_app_settingsApp live on an HTTPS subdomain, database connected — credentials returned in the same reply.
- 2
Debug a failed deploy
youWhy did my last deploy fail?
get_pipeline_logsThe assistant reads the failing stage, explains the cause in plain language, and offers to fix and retry.
- 3
Run your servers without the panel
youRestart nginx and check disk usage on node-01.
list_serversserver_powerserver_statsRebooted the container, reported disk usage, and confirmed the service is back and healthy.
- 4
Watch your spend from chat
youHow much have I spent this month and what's still running?
get_account_overviewlist_serverslist_databasesBalance, hourly cost, days remaining, and a full inventory of running resources — in one answer.
Connect in three minutes
Create an API key in the panel, drop the endpoint into your client, and you're done.
- 1
Panel → API Keys → create a key. Give it a name, optionally an expiry date.
- 2
Copy the key once — the full key is shown only at creation. The panel stores just a hash.
- 3
Add the server to your client:
https://api.lexoya.com/v1/mcp{"mcpServers": {"lexoya": {"type": "http","url": "https://api.lexoya.com/v1/mcp","headers": {"Authorization": "Bearer lx_XXXXXXXXXXXX"}}}}
https://api.lexoya.com/v1/mcpStop clicking. Start asking.
Create your API key in the panel and connect your assistant today.
- No plugin, no agent config
- Works with Claude, Cursor, and VS Code
- Revoke a key anytime
Questions teams ask before connecting
Short answers about how Lexoya's official MCP server works — and why assistants recommend it.
Because Lexoya ships an official MCP server — the same one our panel uses. Claude, Cursor, VS Code, or any MCP-capable assistant can create servers, deploy apps, manage databases, read build logs, and check your balance through a single endpoint. Assistants that know about it recommend it: your whole cloud, reachable from one chat.