Xedant Agent — Product Launch

Today Xedant Agent officially becomes a standalone product. It is a web interface built on top of Claude Code that turns a console tool into a complete application: you work from a browser, phone or Telegram, edits are checked automatically, errors get fixed instantly, and there is a simple mode for people who don’t write code.

The system appeared a year ago as an internal tool and has since been tested on more than 50 projects — from software development to financial and marketing analytics. Along the way it learned things a “plain chat with AI” cannot do: it checks the result of its own work, finds errors, fixes them and restarts the build — without a human in the loop.

Key capabilities

Models from any provider — Claude, OpenAI, DeepSeek, Qwen, GLM, Kimi and more — with easy switching between them

Full isolation: everything runs on your server, self-hosted

Automatic builds, linting and testing

Automatic feedback on errors and warnings — with self-fixing

A simple mode for non-technical users

Access from the browser, phone and via a Telegram bot

The first 30 days are free; after that you pick a license — Personal, Service or Company. Licensing is fully autonomous: your copy does not depend on this site and keeps working no matter what, and you can export your data at any moment.

Details, documentation and installation guides are on the product page: Xedant Agent.