gmc
agent
Warn
Health Pass
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Community trust — 11 GitHub stars
Code Warn
- process.env — Environment variable access in action.yml
- fs module — File system access in action.yml
Permissions Pass
- Permissions — No dangerous permissions requested
Purpose
This CLI tool manages parallel Git worktrees designed for simultaneous use by multiple AI agents. It also generates standardized commit messages and branch names by sending staged git diffs to a Large Language Model (LLM).
Security Assessment
Risk: Medium. The tool requires your LLM API key and sends your code's git diffs to external AI providers over the network. The automated warnings regarding environment variable and filesystem access are expected, as the application manages `.env` files and local git directories. No hardcoded secrets or dangerous background permissions were found. However, using any AI commit tool inherently exposes your proprietary source code to third-party AI APIs.
Quality Assessment
The project is modern, actively maintained (last updated today), and properly licensed under the permissive MIT license. It is written in Go, which provides a clean, standalone compiled binary. Community trust is currently low, reflected by the small number of GitHub stars, meaning the tool is relatively new and hasn't yet undergone widespread peer review or large-scale enterprise adoption.
Verdict
Use with caution: the utility is well-structured and safe to run, but developers must be comfortable configuring it to expose their code diffs to external AI APIs.
This CLI tool manages parallel Git worktrees designed for simultaneous use by multiple AI agents. It also generates standardized commit messages and branch names by sending staged git diffs to a Large Language Model (LLM).
Security Assessment
Risk: Medium. The tool requires your LLM API key and sends your code's git diffs to external AI providers over the network. The automated warnings regarding environment variable and filesystem access are expected, as the application manages `.env` files and local git directories. No hardcoded secrets or dangerous background permissions were found. However, using any AI commit tool inherently exposes your proprietary source code to third-party AI APIs.
Quality Assessment
The project is modern, actively maintained (last updated today), and properly licensed under the permissive MIT license. It is written in Go, which provides a clean, standalone compiled binary. Community trust is currently low, reflected by the small number of GitHub stars, meaning the tool is relatively new and hasn't yet undergone widespread peer review or large-scale enterprise adoption.
Verdict
Use with caution: the utility is well-structured and safe to run, but developers must be comfortable configuring it to expose their code diffs to external AI APIs.
Parallel git worktrees for parallel AI agents — plus AI-generated commits.
README.md
gmc
Parallel worktrees for parallel AI agents — plus AI-generated commits.
A Git workflow CLI built for the AI coding era: spin up a worktree per agent, share .env and node_modules across them, and let an LLM write the commit when you're done.
Installation
Homebrew (macOS/Linux)
brew install samzong/tap/gmc
# or
go install github.com/samzong/gmc@latest
Features
| Command | What it does |
|---|---|
| Worktree — parallel AI development | |
gmc wt clone <url> [--upstream <url>] |
Clone as .bare/ + worktree layout, optionally register upstream |
gmc wt add <name> [-b <base>] [--sync] |
New worktree on a new branch |
gmc wt dup [N] [-b <base>] |
Fan out N sibling worktrees for parallel agents |
gmc wt promote <temp> <name> |
Rename a .dup-N branch to a permanent name |
gmc wt list |
List all worktrees in the family |
gmc wt switch |
Interactive switch between worktrees |
gmc wt remove <name> [-D] |
Delete worktree (and optionally its branch) |
gmc wt sync |
Pull the base branch up to date |
gmc wt share add <path> |
Share .env / node_modules / venv across worktrees |
gmc wt pr-review <pr-number> |
Spin up a worktree from a GitHub PR |
gmc wt prune |
Remove worktrees whose branches are merged |
| Commit — AI message generation | |
gmc |
Generate Conventional Commits message from staged diff |
gmc -a [paths...] |
Stage (all or given paths), then commit |
gmc --branch <desc> |
Generate a branch name, switch, then commit |
gmc --issue <N> |
Append (#N) to the subject |
gmc --prompt <text> |
Extra instruction for the LLM |
gmc --dry-run |
Generate but don't commit |
gmc -y / --no-verify / --no-signoff |
Auto-confirm / skip hooks / skip signoff |
| Other | |
gmc tag [-y] |
Suggest and create the next semver tag |
gmc init |
Interactive setup wizard |
gmc config set <key> <value> / gmc config get |
Manage config |
gmc --output json |
Machine-readable output for agents and CI |
gmc completion zsh|bash|fish |
Shell completion |
Config
Config lives at ~/.config/gmc/config.yaml (legacy ~/.gmc.yaml still works; a project-level .gmc.yaml overrides global). Run gmc init for a guided setup, or set fields manually:
gmc config set apibase https://api.openai.com/v1
gmc config set apikey sk-...
gmc config set model gpt-4.1-mini
gmc config set role "Backend Developer"
Custom prompt template: set prompt_template to a YAML file path with {{.Role}}, {{.Files}}, {{.Diff}} variables. See docs/.
Shell completion
gmc completion zsh > ~/.zsh/completions/_gmc
gmc completion bash > ~/.bash_completion.d/gmc
License
This project is licensed under the MIT License - see the LICENSE file for details
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found