lauvsong

lauvsong / claude-soul

Public

Custom Claude Code settings — rules, hooks, skills, and scheduled tasks

17
0
100% credibility
Found Mar 22, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A backup and security configuration kit for Claude Code that installs protective rules to block dangerous commands and file accesses.

How It Works

1
🔍 Discover claude-soul

You hear about a handy safety kit that backs up your custom settings for your AI coding assistant and keeps it secure from mistakes.

2
📥 Download the kit

You grab the safety kit folder to your computer in a simple way.

3
🚀 Easy setup

Run the quick installer – it copies everything to the right place and safely saves your old settings as a backup.

4
🔒 Add your private details

Create a hidden note with your personal access info so nothing sensitive gets shared.

5
🛡️ Set your safety rules

Pick and add rules to block dangerous actions like deleting important files or running risky commands.

6
✨ Use built-in helpers

Enjoy commands to backup settings, review security, or hand off work without worries.

✅ Fully protected setup

Your AI coding world is now backed up, secure, and ready for safe, smooth use every day.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is claude-soul?

claude-soul is a Python-powered set of custom Claude Code settings you install into ~/.claude for claude ai custom workflows on claude soul github. It backs up your existing configs and deploys security guardrails, rules, skills like /ask and /claude-security-review, plus commands such as /backup-customs and /handoff. Developers get a "soul backup" against losing custom claude code commands, custom claude skills, and accidental destructive actions via AI-generated Bash or file ops.

Why is it gaining traction?

Its security-first approach blocks risky moves like git push, rm -rf, kubectl delete, or sudo without pausing for confirmation—pure allow/deny for smooth sessions. Unified protection covers Bash, file reads/writes/edits, and secrets isolation via local JSON, dodging Claude's buggy settings.json denies. Custom claude agents shine with skills for security audits, GitHub CLI guides, and lesson logging, saving token burn on safe queries.

Who should use this?

DevOps engineers scripting infra with Claude Code, where one bad AI suggestion could nuke clusters or leak tokens. Backend teams building custom claude code plugins or custom claude md rules to harden AI assistance. Solo devs handling sensitive .env, SSH keys, or K8s, tired of babysitting AI outputs.

Verdict

Grab it if you're deep in Claude Code and crave ironclad guardrails—install.sh makes setup dead simple with auto-backups. Low 1.0% credibility score and 17 stars signal early maturity, but claude soul doc quality and focused scope make it a low-risk experiment over vanilla settings.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.