Agent skill

generating-commits

Generates Conventional Commits messages, then commits changes. Use when the user says "commit", "git commit", or asks to commit changes, wants to create a commit, or when work is complete and ready to commit.

Stars 0
Forks 0

Install this agent skill to your Project

npx add-skill https://github.com/sabertazimi/skills/tree/main/skills/generating-commits

SKILL.md

Generating Commits

Generate Conventional Commits messages and commit changes.

When to Use This Skill

Activate this skill when:

  • The user types "commit" or "git commit" (with or without slash command)
  • The user says "commit this" or "let's commit"
  • The user asks to create a commit message
  • Work is complete and ready to commit
  • The user mentions committing or pushing changes

Critical Rules

MUST NEVER add co-author or mention Claude Code/Cursor (ANY agents) in commit messages

Workflow

1. Check Project Preferences

Read CLAUDE.md for commit preferences. Priority: Project preferences override default Conventional Commits. Look for sections mentioning:

  • Commit format/style guidelines
  • Type values (allowed types and their meanings)
  • Scope requirements (required/optional/omitted)
  • Body requirements (required/optional/omitted)

If no preferences defined in the project, fall back to standard Conventional Commits.

2. Gather Context

Collect information about the current git state:

bash
# Current git status
git status

# Current git diff (staged changes)
git diff --staged

# Recent commits for context
git log --oneline -10

# Current branch
git branch --show-current

Edge case: If git diff --staged output is empty (no staged changes), prompt the user to stage files first using git add <files>. Do not proceed with commit until files are staged.

3. Generate Message Candidates

Analyze the diff content to understand the nature and purpose of the changes. Generate 3 commit message candidates based on the changes:

  • Each candidate should be concise, clear, and capture the essence of the changes
  • Follow Conventional Commits format, with project preferences taking priority (type, scope, body requirements)

Format:

gitcommit
type(scope): concise subject line describing what changed

[Summary of the modifications]

4. Execute Commit

IMPORTANT: Do not use git add -A or git add . Commit only the files that are already staged and understood.

Select best candidate, explain reasoning of your choice, then commit with heredoc (for multi-line messages):

bash
git commit -m "$(cat <<'EOF'
type(scope): subject line

[modifications summary]
EOF
)"

Important Notes

  • Use heredoc for multi-line commits - Ensures proper formatting
  • Be specific in summaries
  • Think about the reader - someone explaining this code repository in 6 months
  • No co-authors - Never add "Co-Authored-By" or mention Claude Code/Cursor (ANY agents)

Expand your agent's capabilities with these related and highly-rated skills.

davila7/claude-code-templates

verl-rl-training

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.

23,776 2,298
Explore
davila7/claude-code-templates

openrlhf-training

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.

23,776 2,298
Explore
davila7/claude-code-templates

gguf-quantization

GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.

23,776 2,298
Explore
davila7/claude-code-templates

Claude Code Guide

Master guide for using Claude Code effectively. Includes configuration templates, prompting strategies "Thinking" keywords, debugging techniques, and best practices for interacting with the agent.

23,776 2,298
Explore
davila7/claude-code-templates

qdrant-vector-search

High-performance vector similarity search engine for RAG and semantic search. Use when building production RAG systems requiring fast nearest neighbor search, hybrid search with filtering, or scalable vector storage with Rust-powered performance.

23,776 2,298
Explore
davila7/claude-code-templates

behavioral-modes

AI operational modes (brainstorm, implement, debug, review, teach, ship, orchestrate). Use to adapt behavior based on task type.

23,776 2,298
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results