Commit changes in atomic details, split commits

PromptadminResource

A structured system prompt designed to help LLMs analyse git diffs or modified files and divide them into atomic, logical commits adhering to Conventional Commits standards. The resulting output is structured as an executable Bash script.

Key points

  1. Role prompt: Instructs the AI to act as a Senior Software Architect and Release Engineer specialised in Git best practices.
  2. Atomic grouping: Mandates splitting unrelated modifications (such as UI updates and database schemas) into distinct functional chapters.
  3. Conventional Commits compliance: Enforces standard commit types (feat, fix, refactor, etc.) and scopes written in imperative mood.
  4. Senior-level descriptions: Requires detailed bullet points detailing why and how changes were made, including specific class and function names.
  5. Standardised output: Delivers a single Bash code block containing the exact `git add` and `git commit` commands.

Why it matters

Large, tangled commits complicate code reviews and make git-bisect operations tedious. Automating the division of work into atomic commits improves repository cleanliness and speeds up development workflows.

Analysis

The prompt relies on established prompting techniques: persona/role definition, clear structural breakdowns, negative constraints, and exemplar formatting (few-shot elements for commit conventions). By constraining the output to a clean Bash script block, the prompt offers an efficient utility for developers, DevOps engineers, and technical leads seeking to maintain strict version control standards without manual overhead.

Prompt

# Role
Act as a Senior Software Architect and Release Engineer. You are an expert in Git best practices, clean code, and technical documentation.

# Context
I will provide you with a set of code changes (a `git diff`, a list of modified files, or the raw code snippets).

# Your Objective
Your goal is to break these changes down into logical, atomic "chapters." You must separate unrelated changes into distinct commits rather than lumping them into one. For each atomic change, you will generate the specific Git commands to stage and commit the files.

# Guidelines for Commits

1.  **Atomic Grouping:** Analyze the changes. If I modified the UI logic *and* the backend database schema, these must be two separate commits. Group files by their functional "chapter."
2.  **Conventional Commits Standard:** You must use the Conventional Commits format for the Subject Line:
    `type(scope): subject`
    * **Types:** `feat`, `fix`, `docs`, `style`, `refactor`, `perf`, `test`, `chore`, `build`, `ci`.
    * **Scope:** The specific module or file affected (e.g., `auth`, `navbar`, `api`).
3.  **The Subject Line (Title):**
    * Must be imperative ("Add" not "Added").
    * Must be incredibly specific and descriptive. Avoid vague terms like "update code."
    * Example: `feat(payment): implement stripe webhook handler for recurring billing`
4.  **The Description (Body):**
    * Write for **Senior Developers/Techies**.
    * Use a bulleted list (`-`).
    * Reference specific function names, class names, variables, or logic patterns changed.
    * Explain the **WHY** (motivation) and the **HOW** (technical implementation).
    * Mention side effects or architectural decisions if relevant.

# Output Format
Provide a **single Bash script block** containing the commands. Use the following structure for every logical group:

```bash
# [Chapter Name/Topic]
git add <specific_file_path> <another_file_path>
git commit -m "<type>(<scope>): <subject_line>" -m "<Detailed technical description bullet 1>" -m "<Detailed technical description bullet 2>"

Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.

Similar in content, found by the AI search.

  • Repository:coleam00/skills

    Cole's AI Skills: Modular Skill Files for Coding Agents

    The GitHub repository 'coleam00/skills' provides a collection of 33 practical skills designed for AI coding agents such as Claude Code. Instead of maintaining a monolithic 2,000-line CLAUDE.md instruction file, each skill resides in a standalone folder with a SKILL.md file and is loaded on demand. The set revolves around a standard development cycle: prime, plan, implement, validate, review, commit, and PR.

    482starsPython

    AI & LLMs· Tool

  • Repository:GonChen/real-engineering-prompt

    Real Engineering Prompt: Minimal Behavioural Directives for Coding Agents

    This open-source project offers a concise instruction template (AGENTS.md) designed to bind modern coding agents to core engineering practices like minimal complexity, scoped modifications, and evidence-backed verification.

    4stars

    AI & LLMs· Resource

  • X post:Luke Parker

    Streamlining Pull Request Bodies with Luke Parker's 'writing-pr' Skill

    Luke Parker shared a prompt skill configuration titled 'writing-pr' designed for AI agents and hosted in an agents.md file. The skill aims to generate concise and readable pull request descriptions rather than verbose text.

    1.2kbookmarks34.8kviews

    AI & LLMs· Resource

  • Repository:davidsolheim/skills

    davidsolheim/skills: Reusable Grok Agent Workflows for Linear and Git

    This repository provides portable agent skills for xAI's Grok designed to handle the end-to-end product delivery cycle from Linear ticket creation to merged pull requests. The skills are sanitized to exclude hard-coded local paths, client names, and organization-specific boards.

    1stars

    AI & LLMs· Resource

  • Repository:mattpocock/skills

    mattpocock/skills: Modular Agent Skills for Real Software Engineering

    Matt Pocock shares a curated collection of agent skills taken directly from his own development setup. The project positions itself against rigid frameworks like GSD, BMAD, and Spec-Kit, as well as uncontrolled 'vibe coding'. Instead, it provides small, composable, and adaptable skills compatible with Claude Code, Codex, and other coding agents.

    247.2kstarsShell

    AI & LLMs· Tool

Let's talk about your project

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.