# Commit Staged

> Create a commit from already staged changes, using a supplied message or a concise conventional commit message inferred from the diff.

- **Type:** Skill
- **Install:** `agentstack add skill-jtsternberg-claude-plugins-commit-staged`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [jtsternberg](https://agentstack.voostack.com/s/jtsternberg)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [jtsternberg](https://github.com/jtsternberg)
- **Source:** https://github.com/jtsternberg/claude-plugins/tree/main/plugins/git-commits/skills/commit-staged

## Install

```sh
agentstack add skill-jtsternberg-claude-plugins-commit-staged
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Commit Staged Changes

Use this skill only when the user explicitly asks to commit their staged changes.

**Arguments provided:** $ARGUMENTS

Codex: if the invocation text above is not populated, use a commit message supplied after the skill name or in the current request. If none is available, infer the message from the diff.

## Workflow

1. Inspect the repository state with `git status`, `git diff`, and `git diff --cached`.
2. Stop if there are no staged changes. Do not create an empty commit.
3. Respect a commit message supplied by the user. Otherwise, derive a concise conventional commit message from the staged diff and the repository's recent commit style (`git log`).
4. Check that the staged files do not include secrets or private configuration (for example `.env`, credential files, tokens, or private keys). Stop and tell the user if they do.
5. Create the commit from the staged files. Do not amend, use `--no-verify`, or push unless the user explicitly asks.
6. Report the commit hash and the remaining `git status` state.

If the staged diff contains unrelated changes, stop before committing and ask the user to split or clarify the intended scope.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [jtsternberg](https://github.com/jtsternberg)
- **Source:** [jtsternberg/claude-plugins](https://github.com/jtsternberg/claude-plugins)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** yes
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-jtsternberg-claude-plugins-commit-staged
- Seller: https://agentstack.voostack.com/s/jtsternberg
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.