Vibe Safe — AI skill for Claude Code
Active session guardian for non-technical contributors shipping AI-assisted code in shared codebases.
How to install Vibe Safe
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open googlarz/vibe-safe and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Vibe Safe does
Active session guardian for non-technical contributors shipping AI-assisted code in shared codebases.
Alternatives in AI
- Skill Seekers Roadmap — Transform Skill Seekers into the easiest way to create Claude AI skills from any knowledge source - documentat 11.1k ★
- Cartographer — Claude Code plugin that maps and documents codebases of any size using parallel AI subagents 518 ★
- Aelf-skills Contribution Guide (Adding New Skills) — This guide is the hard gate for both human contributors and AI agents 182 ★
README
vibe-safe
**For the non-technical contributor:** Claude wrote the code. You read the first two lines, it looked fine. The commit message says "update." You accepted "theirs" on all the merge conflicts. The test suite is green because Claude lowered the coverage threshold, not because the tests pass. The API key is in a file you never opened — it won't appear in your diff.
Run this before you commit. It reads your actual git state, not what you assume is in it.
**For the developer:** Everything above has already happened in your repo. Install this before it happens again.
**40 risk categories. Developer-defined contracts. Every flag cites file:line. Claude reads the actual git state — nothing is self-reported.**
What it guards against
| Risk | Without vibe-safe | With vibe-safe |
|---|---|---|
| Credential committed (not in diff) | Missed — not in staged files | Caught — full-repo grep, src/api/client.ts:14 |
| Committing directly to main | No check | Stopped — feature branch created automatically |
| Config/auth/migration/quality-gate files in diff | No check | Flagged with escalation instruction |
| Merging conflict by accepting all "theirs" | No check | Both sides explained, Danger Zone → stop |
| Claude lowered a quality threshold to pass CI | No check | Caught — diff scanned for numeric decreases in quality configs |
@ts-ignore / eslint-disable added |
No check | Flagged — suppressed error, not fixed |
Failing test skipped with .skip() |
No check | Flagged — bypassed test, not fixed |
| Test file deleted | No check | Flagged — suite passes by omission |
console.log / debugger left in code |
No check | Flagged — debug artifact in production |
Empty catch {} block |
No check | Flagged — errors silently swallowed |
Lock file changed without package.json |
No check | Flagged — undocumented dependency change |
| Binary file committed | No check | Flagged — permanent history bloat |
| PII in |
Related Skills
Explain It With Pictures
A skill for AI coding agents (Claude Code, opencode, Cursor): explain any codebase to non-technical builders w
AI Engineering Playbook
Practical playbook for AI-assisted engineering: rules, role separation, and guardrails for teams using Cursor/
Agentic Engineering Methodology
Battle-tested 7-phase methodology for shipping real software with AI coding agents (Claude Code, Cursor, Copil
Claude Code Skills
Complete software development lifecycle skills optimized for non-technical founders building SaaS applications
OS Coach
A Claude Code skill that coaches a non-technical person through building their own agentic OS, one layer at a
AI Business Builder Agents
A multi-agent plugin for Claude Code that helps non-technical founders build AI-powered businesses from idea t
Related Agents
Slop Detector
Adversarial generic-design hunt with fresh context — finds structural, component, visual, copy, interaction, a
Po Guardian
Use when validating contractual outcome (C2), ICP fit, scope decisions, or before approving promotion to ASSIS
Principal
Use this agent for technical design or architecture review involving APIs, storage, concurrency, process owner