X-AIApp is here: describe the change, review it, ship it
Our AI application builder turns a task written in plain language into a reviewed branch and a live deploy, with a person deciding at every gate.
Most teams don't need another code assistant. They need a way to get a change from "we should…" to running in production without losing control of the code, the servers or the keys. That is what X-AIApp does.
From a sentence to a running service
- Describe the task. Write the application or change in plain language. If you only have a question, ask it, and the answer appears on the task.
- An agent works on your own worker. A coding worker runs the agent on the repository and pushes a task branch. Workers only make outbound connections, so they can sit behind NAT or a VPN.
- Checks run, and fix themselves. After coding, the project's check command runs. If it fails, the agent gets the output and fixes the problem before anyone reviews it.
- A person reviews. Read the diff, send it back with an instruction, or merge. A change whose checks failed can't be merged without a written reason.
- Deploy. Upload, build, switch, restart and wait for the health check. If activation or the check fails, the previous release comes back.
AI where it helps, automation where it matters
The model writes code. It never touches production.
- Deploys never use an LLM. Every deploy step is fixed automation you can read and predict.
- Only merged work reaches servers. Workers can push only their own task branch, and those branches can never deploy by themselves.
- Pick the model per task. X-AIApp uses the task's model, then the project default, then the global default. Every run records its model, tokens and cost.
Code, docs and servers in one place
Each project has hosted Git with a code browser, a Docs tab for requirements and decisions, and a wiki. Register a server once, and a fixed, read-only probe reports its OS, runtimes, databases, nginx sites, certificates and ports.
Your code, your servers, your keys
Workers see a project only while they run one of its tasks. Server and model secrets are encrypted with AES-256-GCM, host keys are pinned on first connect, and model keys reach a worker only inside the job that needs them.
Try it
Bring one repository and one server, and we'll take your first task to a live deploy with you. See X-AIApp or talk to us.
