#1616th of 22 in Code review and testing
codegpt
Go CLI that writes git commit messages and code reviews from diffs
- Stars
- 1.5k
- License
- MIT
- Last commit
- Sep 2026
- Last release
- Sep 2026
- Language
- Go
1,483 on 3 Apr 20261,504 on 30 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
CodeGPT is a Go command-line tool that reads your git diff and generates conventional-commit messages or short code review summaries. It installs a git prepare-commit-msg hook and talks to OpenAI, Azure OpenAI, Gemini, Anthropic, Groq, Ollama, OpenRouter or a LiteLLM proxy. Prompts are editable templates, and API keys can come from the OS keyring or a shell helper command.
Who it is for: Developers who want LLM-written commit messages and quick reviews in git
Strengths
- Supports eight provider options, including local Ollama and a LiteLLM proxy
- Installs a prepare-commit-msg hook so messages are generated during git commit
- Prompt templates are editable files, loaded with `codegpt prompt --load`
- API key helper command pulls keys from 1Password, AWS or GCP secret stores
Weaknesses
- Only commit messages and brief reviews; it does not edit code or run commands
- Default max_tokens is 300, which can truncate longer outputs
- Output language options are limited (en, zh-tw, zh-cn, ja)
- No Docker image; install is by script, binary, Homebrew, Chocolatey or go install
What it needs
- no GPU
- Docker
- Needs git, API key for an LLM provider, or a local Ollama server
- Models: OpenAI (gpt-5, gpt-4o, gpt-4), Azure OpenAI, Gemini, Anthropic, Ollama
Quick answers
Is codegpt open source?
Yes. Its code is on GitHub under the MIT license.
Can you run codegpt with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Also in Code review and testing
See all 22| Rank | Project | Score |
|---|---|---|
| 1 | playwrightCross-browser automation and end-to-end test framework for Chromium, Firefox and WebKit | 94 out of 100 |
| 2 | open-code-reviewCLI that reviews Git diffs with an LLM agent and line-level comments | 91 out of 100 |
| 3 | react-doctorCLI that scans React codebases for state, performance, security and accessibility issues | 89 out of 100 |
| #4 | pr-agentLLM pull request reviewer that comments via /review, /describe and /improve | 86 out of 100 |
| #5 | e2eAI-driven end-to-end testing framework for web and mobile apps | 86 out of 100 |