vibe-security-skill — Claude Code skill: security audit
Audits code written with an AI assistant for the usual holes: exposed keys, broken RLS policies, weak auth.
npx skills add reason-machines/security-skills --skill vibe-security-skill -a claude-codeWhat vibe-security-skill does
The skill scans a codebase for vulnerability patterns that AI coding assistants tend to leave behind. It ships technology-specific reference files, so only the parts of your stack that actually exist get checked: Supabase, Stripe, React Native and others.
What it catches: secrets and API keys sitting in source, Row-Level Security that is off or missing, insecure auth flows, payment amounts submitted by the client, no rate limiting, tokens kept in localStorage, secrets leaking into mobile bundles, AI API keys with no usage cap.
It activates on its own while you work on auth, payments, database queries, API key setup, user data handling and environment variables. In Claude you can also call it directly with /vibe-security. This is a pattern check, not a full architecture review.
What it helps with
- find secrets left in the source tree
- confirm RLS is on and policies are not wide open
- make sure the server sets the payment amount, not the client
- catch tokens stored in
localStorage - check rate limits on expensive AI endpoints
Example prompts for the agent
- Audit my code for security issues
- Is this payment flow secure?
- Review my Supabase RLS policies
How to install vibe-security-skill
1Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill vibe-security-skill -a claude-code2Manually: put the files in place
~/.claude/skills/vibe-security-skill/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/vibe-security-skill/
1Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill vibe-security-skill -a codex2Manually: put the files in place
.agents/skills/vibe-security-skill/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill vibe-security-skill -a cursor2Manually: put the files in place
.cursor/skills/vibe-security-skill/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install reason-machines/security-skills vibe-security-skill2Manually: put the files in place
.github/skills/vibe-security-skill/SKILL.md1Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill vibe-security-skill -a gemini-cli2Platform CLI
gemini skills install https://github.com/reason-machines/security-skills --path skills/vibe-security-skillManually: put the files in place
.devin/skills/vibe-security-skill/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill vibe-security-skill -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about vibe-security-skill
How do I install vibe-security-skill in Claude Code?
Run npx skills add reason-machines/security-skills --skill vibe-security-skill -a claude-code. The skill lands at ~/.claude/skills/vibe-security-skill/SKILL.md.
What does vibe-security-skill do?
It looks for common vulnerabilities in apps built with an AI assistant: exposed secrets, RLS gaps, weak authentication, unsafe payment handling.
When does the skill activate on its own?
While you work on authentication, payments, database queries, API key configuration, user data or environment variables. In Claude you can also trigger it with /vibe-security.