⚡ AI-Git-Bot
The self-hosted AI workflow automation platform for Git repositories. AI-Git-Bot automates the engineering chores every team knows should happen — PR reviews, tests, docs, translations, issue refinement, coding — triggered by the Git events you're already producing. No new process. No migration. No vendor lock-in.
Then open http://localhost:8080 — create your admin account and wire the first bot
Every engineering team knows what should happen — careful PR reviews, regression tests for bugs, up-to-date documentation, consistent translation coverage. The problem isn't disagreement. It's that these tasks are uncomfortable, repetitive, and easy to postpone when deadlines get tight. AI-Git-Bot turns them into repeatable workflows that happen automatically.
No new development process
The bot works inside your existing Git platform. Your developers don't need to learn a new tool.
No migration project
Connect to your existing repositories via webhooks. Nothing needs to move.
No vendor lock-in
Mix any AI provider with any Git platform. Self-hostable end-to-end. Your code stays in your infrastructure.
Workflows — automation, not another coding assistant
AI-Git-Bot is not another tool to help developers write code faster. It's a self-hosted automation layer for the work around the code — every workflow below is triggered by an event your team already produces, and each one ships behind an opt-in flag.
Who is AI-Git-Bot for?
Three reasons teams adopt it — many teams overlap across these concerns.
🏢 Running Gitea and missing modern AI tooling?
Many teams choose Gitea for ownership of their source code. Unfortunately, most AI products focus on GitHub. AI-Git-Bot brings AI code reviews, test generation, issue authoring, documentation upkeep and interactive PR discussions directly into Gitea — without forcing developers to leave their existing workflows.
"Copilot-style workflow automation for Gitea."🔒 Need AI but cannot send source code to external services?
Many organizations cannot use cloud-only AI products due to compliance, privacy or contractual requirements. AI-Git-Bot supports Ollama, llama.cpp, self-hosted Git platforms, private networks and provider-independent architectures. Source code, prompts, credentials and models remain under your control.
"AI automation without handing your repos to a SaaS vendor."🚀 Maintaining repositories with too much engineering overhead?
Every repository accumulates engineering chores: reviews, tests, documentation, acceptance criteria, follow-up fixes. AI-Git-Bot turns those into repeatable workflows triggered by events your team is already producing — PR opened, reviewer requested, issue assigned, @bot mentioned.
Integrations — mix & match any AI provider with any Git platform
AI-Git-Bot is a gateway, not a script. Every Git platform plugs in through a RepositoryApiClient SPI. Every AI provider plugs in through an AiClient SPI. One admin UI, one PostgreSQL database, one set of encrypted credentials.
🤖 AI Providers
🔗 Git Platforms
Screenshots across every Git platform
AI-Git-Bot works the same way everywhere — reviews, inline comments, E2E test reports, and coding agents across Gitea, GitHub, GitLab, and Bitbucket.
Getting started — up and running in 5 minutes
One Docker image, one PostgreSQL database, one admin UI. No Kubernetes required.
Prefer not to assemble the YAML by hand?
Generate your instance — choose AI provider, Git platform and
workflows, and get a ready docker-compose.yml plus the exact webhook to paste into your
repository. The generator runs entirely in your browser; no key or token is ever sent to this site.
Run the container
docker run -p 8080:8080 tmseidel/ai-git-bot:latest
Open the Admin UI
Navigate to http://localhost:8080 and create your administrator account.
Connect AI & Git
Create an AI Integration (Anthropic, OpenAI, Ollama, …) and a Git Integration (Gitea, GitHub, GitLab, Bitbucket). API keys are encrypted at rest (AES-256-GCM).
Create a Bot
Create a bot combining one AI integration with one Git integration. Enable the workflows you want. Copy its unique Webhook URL.
Wire the Webhook
Configure the webhook in your Git provider. Done. Assign the bot as a reviewer on a PR — watch it post its first review.
Designed for production from day one
Adoption risk is the #1 reason AI-in-the-SDLC pilots die. Every design choice below exists to keep that risk near zero.
Opt-In Everywhere
Every workflow ships behind a prworkflow.<name>.enabled flag. Only review defaults to true. Nothing changes for repos you don't touch.
Encrypted at Rest
Every credential, webhook secret, and deployment-target config is encrypted with AES-256-GCM. API keys never appear in logs.
Operator-Owned Tool Whitelists
Per-bot built-in tool whitelist + per-MCP-config tool whitelist. The admin decides what the agent can do — not the LLM, not the prompt.
No Silent Deploys
The bot triggers your existing CI — it never invents a deploy path. Four interchangeable DeploymentStrategy implementations keep deployment under your control.
Fully Auditable
Every workflow run, step, and tool call is persisted. Micrometer meters (prworkflow.run_total, prworkflow.run_duration_seconds) exposed at /actuator/prometheus.
Cost-Bounded
Workflow params expose hard caps (maxTestCases, maxRetries) and the agent loop honours the shared agent.budget.* token budget.
Suite Promotion
Generated E2E tests don't evaporate on PR close. Four lifecycle modes (ephemeral, commit-to-pr, offer-as-pr, promote-on-merge) let tests graduate into the codebase under operator control.
Backwards Compatible
Every new column on bots is nullable. Existing bots without a WorkflowConfiguration keep running the legacy review workflow only. No migration drama.
News — what changed lately
The three latest entries from the news feed: release notes and engineering write-ups, one markdown file each. The complete list, an Atom feed and every entry in full live at /articles/.
Not another coding assistant
It's a self-hosted automation layer for software delivery workflows. If your team already knows what good engineering practices look like — but struggles to do them consistently — AI-Git-Bot was built for exactly that problem. Wire one bot. Let the chores take care of themselves.
Commercial services — adopt it without carrying the operations work
AI-Git-Bot is MIT-licensed today and stays MIT-licensed. If you want it running in your organisation without spending engineering weeks on it, three packages are available on top of the open-source project: consulting, setup and hosting — either with your own provider keys (BYOK) or with a self-hosted LLM.
Consulting
Find out which workflows actually pay off in your repositories — and how to configure them so the output is reviewable instead of noisy.
- Workflow shortlist for your repos and team size
- Prompt, model and cost-budget tuning on real PRs
- Provider decision: cloud API, BYOK or local LLM
- Rollout plan, guardrails and team onboarding
Setup & integration
AI-Git-Bot installed on your infrastructure and wired to your Git platform and AI provider — including an acceptance run on a real repository and a handover you can operate yourself.
- Container deployment behind your proxy and TLS
- Webhooks for Gitea, GitHub, GitLab or Bitbucket
- AI provider wiring — API key or self-hosted LLM
- Acceptance run on a live repo + written handover
Hosting
Managed operation of a dedicated instance. With BYOK your keys, billing and rate limits stay yours; with a self-hosted LLM the inference runs in the deployment, so no source code leaves it.
- Updates, PostgreSQL backups, monitoring and alerts
- BYOK: encrypted at rest, visible only to you
- Self-hosted inference: Ollama / llama.cpp, air-gap capable
- Response times and support channel agreed in writing
What comes with every engagement
Small ramp-up
One container, no new process, no migration. The first bot reviews a real PR in the first week — the effort goes into configuration, not a project plan.
Professional run & maintenance
Version upgrades, database backups, certificate renewal, log and metrics review — plus release notes filtered to what actually affects your deployment.
Exclusive extensions
Custom workflows, prompt packs and SPI adapters for your in-house systems — built for you, kept private, and not part of the open-source release.
Security & compliance review
Data-flow review for providers and repositories, credential handling (AES-256-GCM), and documentation you can hand to your DPO or auditor.
Knowledge transfer
Hands-on sessions with the maintainer, so your team can add workflows and providers — and switch to running it alone whenever it makes sense.
No lock-in
The software stays MIT, self-hostable and yours. Every engagement can end without the instance ending with it.
Request a quote
Describe what you run today — Git platform, number of repositories, and whether you want to use your own API keys or a local model. You'll get an answer with a proposed scope and a price.