v2.2 — scanner + review call

Scan your vibe-coded app.
Then get a human to finish it.

30-point security scan in seconds. Hit a wall? Book a 30-min review with a senior engineer — free during beta — and walk away with a build plan.

⌨
Go deeper with the full CLI
The web scanner catches common patterns. The full skill runs framework-specific detection, deeper analysis, and generates fix-ready output.
📡
AI builds fast. Learn what it misses.
Practical guides on using AI in business without shipping security holes or burning trust.
Read on Substack →
howll.ai — AI for business →
joshhorneman.com →
Share your score

Built by howll.ai · independent tools for the Claude Code community · open-source & MIT-licensed

30
security checks
5
compliance frameworks
20–40
typical first score · grade F
// the vibe-coding journey — where a-vibe-check fits
💡
Idea
AI helps start
🛠
Build
AI ships fast
🧱
Stuck
you're here
🔍
Scan
find what's broken
🤝
Ship
match with a human

30 checks · 10 categories

Every check targets a vulnerability pattern AI code generators introduce. Each includes detection commands and fix instructions.

ASecrets & Credentials
  • API keys in frontend code
  • .env files in git history
  • Weak JWT secrets
BAuth & Sessions
  • No rate limiting on auth
  • JWTs in localStorage
  • Tokens that never expire
  • Sessions not invalidated
  • Weak password hashing
CAuthorisation
  • Frontend-only admin auth
  • Missing auth middleware
  • IDOR vulnerabilities
DInjection & Input
  • SQL injection
  • XSS via unsafe rendering
  • Missing input validation
  • eval / exec / pickle
EInfrastructure
  • CORS wildcard
  • Running as root
  • Database port exposed
  • No HTTPS enforcement
  • Debug mode in production
FError Handling
  • Stack traces in responses
GBaaS & Third-Party
  • Supabase/Firebase RLS off
  • Unsigned payment webhooks
  • Hallucinated dependencies
HFile Handling
  • Uploads without MIME check
ITransport & Headers
  • Open redirects
  • Missing security headers
  • CSP missing or permissive
JOps Security
  • No logging or monitoring
  • No backup or recovery

Every check weighted by severity

Your score is the percentage of applicable points you pass. N/A checks excluded.

10
Critical
10 pts · 12 checks
7
High
7 pts · 10 checks
4
Medium
4 pts · 7 checks
2
Low
2 pts · 1 check

F <60 · D 60–69 · C 70–79 · B 80–89 · A 90–100 · max: 220 points

Maps to the frameworks that matter

Every failed check links to specific clauses. Turns a technical audit into a business risk conversation.

Privacy Act (APP 11)OWASP Top 10SOC 2 Type IIPCI DSS v4.0GDPR Art. 32

If you're an Australian business, APP 11 requires "reasonable steps" to protect personal information. These 30 checks are free. Hard to argue they're not reasonable.

Three ways your AI fails you.
All of them have a human fix.

You didn't buy Claude. You rented it. When it loops, burns tokens, or leads you somewhere it can't get you out of — that's when the marketplace earns its keep.

scenario 01
Claude's stuck in a loop.
It's rewriting the same file for the fifth time. Each pass is confidently wrong in a slightly different way. You know something's off but you can't see what you can't see.
What helps
scenario 02
You blew the token budget.
Three weeks in, the monthly plan is gone and the fix is further away than when you started. You need a direction, not more tokens.
What helps
scenario 03
Claude pushed you down a path.
Wrong framework, wrong data model, wrong architecture — and now it's three sprints deep. You can see the mess but you can't see a way back out alone.
What helps

The scan finds the wall.
A human gets you over it.

AI gets you 80% of the way. Then it loops. Then it burns tokens. Then it ships you something you don't trust. A Vibe Check connects vibe coders with humans who can finish the job — audits, scaling, design, launch, any of it.

// start here

Book a 30-min vibe-check review.

A senior engineer walks your repo with you. Finds the gaps AI missed. You leave with a short written build plan — and a clear view of what to ship, fix, or rebuild next.

Free during beta ~30 min No card required
Prefer async?
Request a specialist match.
Tell us the problem. We hand-pick 1–3 specialists within 48 hours with a fixed-scope quote before any work starts.
For experts
Developers, designers, strategists — join the pool.
Get paired with paying projects from vibe coders who've already scanned their repo and know where they're stuck. Pick the work you want.
🔒Security & Compliance
Real audits, penetration tests, secure-by-design reviews. GDPR, SOC 2, PCI DSS, Privacy Act mapping. Fix the issues A Vibe Check found.
pen-test · audit · RLS · auth
📈Architecture & Scaling
Your app works at 10 users. Will it work at 10,000? Database design, query optimisation, caching, infra cost reduction, queue design.
postgres · redis · k8s · edge
🎛Frontend & UX Engineering
Accessibility, performance, state management, animation. Make it feel fast, feel right, and work for every user on every device.
react · next · svelte · a11y
⚙Backend & Integrations
API design, Stripe flows, webhook hardening, auth, third-party integrations, background jobs. The plumbing your AI glossed over.
node · python · stripe · oauth
🎨Design & UX
UX audits, design systems, conversion-focused UI, brand and landing pages. Stop shipping Cursor-default-grey. Ship something people want to use.
figma · ux · conversion · brand
🚀Product, Growth & Launch
Positioning, landing pages, launch strategy, pricing, content, SEO. Turn your working prototype into something people find, trust, and pay for.
positioning · launch · seo · copy
Founding expert slots — all 8 archetypes open
0 available this week
For experts Senior developer, designer, or strategist? We're onboarding founding experts to pair with vibe coders who've already scanned their repo.
01
Scan or describe
Run A Vibe Check on your repo — or just tell us where you're stuck. Your scan results pre-fill the brief automatically.
02
We match, not list
No endless profiles to scroll. We hand-pick 1–3 specialists who've solved your exact problem before, with a fixed-scope quote up front.
03
Ship with confidence
Work together in whatever tool fits — Claude Code, GitHub, Slack, a call. Pay on milestones. You own the code.
All experts vetted Fixed-scope quotes Milestone payments You own the code NDAs on request AI-friendly workflow

What happens next.

The call isn't the pitch. It's the plan. Here's what usually comes after a review.

A
You take the plan and run.
The build plan goes into your Claude Code, Cursor, or team — whoever's doing the work. You know what's broken, what to fix first, and what to ignore. No follow-on needed.
Most common
B
We fix the critical stuff in a week.
Some findings are too gnarly for AI to close without human eyes. We scope a fixed-price sprint — usually security hardening, auth fixes, or unblocking a stuck architecture — and ship it.
~1 week · fixed scope
C
We rebuild what's under the vibe.
The call makes it clear the shortest path isn't a patch. The team builds the real thing with you — clean architecture, production-grade auth, tests you can trust, the lot. Claude-friendly so you can still iterate.
multi-sprint · scoped per phase

Paths B and C are the upsell. Path A is the default. We'd rather tell you the plan is trivial and send you on your way than invent work.

Install the full CLI tool

The web scanner above catches the most common patterns. The full CLI skill runs deeper analysis with framework-specific detection.

Claude Code
$ claude install a-vibe-check
Cowork
# download .skill → drag into session

Framework detection: Next.js · React · Express · Django · Flask · Supabase · Firebase · Docker

The scan takes five minutes.
The review takes thirty.
Both are free.

Run the scan. If you're stuck, book the call. Walk away with a plan.