Why does the /commit skill remove Claude attribution?

To ensure commits appear user-authored and avoid automated attribution in the history.

Can I still push after committing?

Yes. After commit, you will be prompted to generate reasoning and may push to the remote branch if requested.

What happens to previous build attempts?

Build attempts are cleared so you start fresh for the next feature, ensuring clean build history.

git-commits

npx machina-cli add skill parcadei/Continuous-Claude-v3/git-commits --openclaw

Files (1)

SKILL.md

1.0 KB

Git Commit Rules

When the user asks to commit, push, or save changes to git:

MUST Use /commit Skill

DO NOT run git commit directly. Instead:

Skill("commit")

The /commit skill:

Removes Claude attribution from commits
Generates reasoning.md capturing what was tried
Clears build attempts for next feature

Why This Matters

Regular git commit adds "Generated with Claude Code" and Co-Author lines
The /commit skill removes these so commits appear user-authored
Reasoning capture preserves build history for future sessions

Trigger Words

When you see these in user prompts, use the commit skill:

"commit", "push", "save changes"
"push to github", "push changes"
"commit and push"

After Commit

The skill will prompt you to run:

bash "$CLAUDE_PROJECT_DIR/.claude/scripts/generate-reasoning.sh" <hash> "<message>"

Then push if requested:

git push origin <branch>

Source

git clone https://github.com/parcadei/Continuous-Claude-v3/blob/main/.claude/skills/git-commits/SKILL.mdView on GitHub

Overview

Git Commit Rules ensure commits are user-authored by routing changes through the /commit skill. It removes Claude attribution, captures a reasoning.md for build history, and resets build attempts so you start fresh for the next feature.

How This Skill Works

Use only the /commit skill for commits, not git commit. The /commit flow removes Claude attribution, creates reasoning.md with what was tried, and clears prior build attempts. After committing, you’ll be prompted to run a reasoning script with the hash and message, and you may push with git push origin <branch> if requested.

When to Use It

When you need to commit changes without Claude attribution so the history shows user authorship.
When you want the system to generate and store a reasoning.md describing what was tried.
When you want to clear previous build attempts before starting a new feature.
When you plan to push changes to a remote repository after commit.
When a user asks to commit, push, or save changes and trigger the /commit flow.

Quick Start

Step 1: When asked to commit, use the /commit skill instead of running git commit directly.
Step 2: If prompted, confirm the commit message and hash, and allow the system to generate reasoning.md.
Step 3: If instructed, run git push origin <branch> to push the changes.

Best Practices

Use the trigger words commit, push, or save changes to invoke the /commit skill.
Review the intended commit message before the /commit flow completes.
Confirm that reasoning.md is generated and that build state is reset.
Avoid direct git commit; rely on the /commit workflow for attribution control.
After commit, follow the generated prompt to run the reasoning script with the correct hash and message.

Example Use Cases

A developer finishes a feature and asks to commit; the system uses /commit to produce a user-authored commit and reasoning.
After commit, the assistant prompts to generate reasoning and then pushes to origin.
A fix is made; the user saves changes; the tool commits via /commit, removing Claude attribution.
A PR is prepared with a clean, user-authored commit history and an accompanying reasoning.md.
Build attempts are cleared automatically after a successful commit to prepare for the next feature.

Frequently Asked Questions

Add this skill to your agents