Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Commit

ASecurity

Selectively stages and commits only the changes related to the current session, skipping unrelated modifications.

477 stars
0 votes
0 copies
0 views
Added 9/22/2026
ai-agentsgitapi

Works with

api

Security Analysis

A100/100

Scanned 9/22/2026

Install to Claude Code

$npx -y skills add greenfield-inc/Pane --skill commit --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Commit?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Commit
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/greenfield-inc-commit/badge)](https://www.skillsdirectory.com/skills/greenfield-inc-commit)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: commit
description: Selectively stages and commits only the changes related to the current session, skipping unrelated modifications.
argument-hint: "[optional: commit message or description of what to commit]"
---

# Commit Agent

Commit only the changes you made in this session to the local branch. Ignore all other changes. **Do not ask for confirmation at any step — just classify, stage, commit, and report.**

## Step 1: Understand What Was Done

Gather context about what you implemented:
1. Check for plans in `./tmp/done-plans/` and `./tmp/ready-plans/` — if any exist, read them for file lists and feature descriptions.
2. If no plans exist, use the conversation history to understand what files you created or modified and why.
3. If `$ARGUMENTS` is provided and does NOT match the `type: description` commit message format, use it as additional context for what should be committed (it will be used for classification in Step 3, not as the commit message).

## Step 2: Inspect All Changes

1. Run `git status` to see all modified, added, and deleted files.
2. Run `git diff` (unstaged) and `git diff --cached` (staged) to see the actual changes. Treat both as a single pool of changes to classify.
3. If there are no changes at all, tell the user there is nothing to commit and stop.

## Step 3: Classify Changes

For each changed file (whether staged or unstaged), determine if it was changed **by you in this session** or not:

**Your changes** — include these:
- Files you explicitly created or edited during this conversation
- Files referenced in plans you implemented (done-plans or ready-plans)
- Supporting changes (imports, types, config) that are clearly tied to your work
- `./tmp/done-plans/` files and `./tmp/context.md` changes associated with your work

**Not your changes** — skip these entirely:
- Files you did not touch in this conversation
- Pre-existing modifications from before the session
- Changes from other agents or manual edits unrelated to your task
- Unrelated `./tmp/` files (research notes, other plans)

When in doubt, include the file rather than leaving it out.

If zero files are classified as yours, tell the user that no changes match this session's work and stop.

**Do not present the classification for confirmation. Proceed directly to staging.**

## Step 4: Stage and Verify

1. `git add <specific files>` — only the files you changed. **Never** `git add .` or `git add -A`.
2. Review the staged diff (`git diff --cached`) for secrets or credentials:
   - API keys, tokens, passwords
   - .env files or credential files
   - Private keys or certificates
3. If secrets are found, **warn the user**, unstage the offending files, and ask how to proceed. Do not commit files containing secrets.

## Step 5: Create the Commit

Do not ask for confirmation. Just create the commit.

1. Write a commit message:
   - If `$ARGUMENTS` matches the `type: description` format (e.g., `feat: add commit skill`), use it verbatim as the commit message.
   - Otherwise, derive a message from the work context.
   - Format: `type: short description` (feat, fix, refactor, docs, chore). Under 72 characters. Imperative mood.
   - Add a body with bullet points if the commit covers multiple logical changes.
2. Create the commit to the local branch. Do not push.

## Step 6: Report

Present the result. Only suggest `/commit` again if there are uncommitted files remaining.

```
Committed: <short sha> <commit message>

Files included:
  - <file list>

Files left uncommitted:
  - <file list, or "none">

Next steps:
  - `/prepare-pr` — Rebase, build, and open a PR
```

Attribution

greenfield-incgreenfield-inc
View sourceMore from greenfield-inc →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1066601 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

651 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →