# AI-Git-Bot

A self-hosted AI workflow automation platform that reviews pull requests, generates tests, triages issues, and keeps documentation and translations in sync, across GitHub, GitLab, Gitea, and Bitbucket.

AI-Git-Bot turns recurring engineering chores that teams agree should happen but rarely get done reliably, such as carefully reviewing every pull request, writing regression tests for bugs, giving issues acceptance criteria, and keeping documentation and translated locale files current, into automated workflows that run inside a team's existing Git platform. It can review pull requests, generate tests, improve issues, triage and route new issues, turn an issue into a pull request, create and run end-to-end tests, keep documentation and translations synced with the code, answer questions inside code reviews, and expose metrics for observing its own workflows.

It is built in Java with Spring Boot and is aimed particularly at teams running Gitea who want GitHub-Copilot-style AI workflow automation without switching platforms, and at organizations that cannot send source code to external cloud AI services: it supports local model runners for fully local inference alongside hosted providers, so source code, prompts, credentials, and models can stay under the team's own control, with no new development process, migration project, or vendor lock-in required.

Website: https://gitbot.eu