FinishKitsecurity
Supabase's October 30 Data API Change: How to Fix New Tables Safely
From October 30, 2026, new Supabase tables are unreachable from supabase-js until you grant access. What breaks, the safe fix, and why RLS comes first.
12 min read
Shipping AI-Built Apps: Guides and Checklists
Practical guides on shipping AI-built apps. From prototype to production, we cover the last mile.
analysis
Why Your AI-Built App Will Fail in Production
AI coding tools are great at prototypes and terrible at production software. Here's why, with real failure modes and what to do about them.
9 min read
tools
Bolt.new Apps: 7 Production Issues and How to Fix Them
Bolt.new gets you to a working app in minutes, but its code has patterns that break in production. Seven specific issues, with code-level fixes.
10 min read
guides
Is My Cursor Project Ready to Ship? The Complete Checklist
It works locally, but is your Cursor project ready for real users? A pre-launch checklist for security, testing, error handling, and deploy config.
7 min read
guides
How to Deploy Your Lovable App to Production (2026)
Your Lovable app works in preview, but preview isn't production. A step-by-step guide to deploying with proper security, config, and monitoring.
9 min read
security
Lovable App Security Checklist: 10 Things to Fix Before Launch
Lovable builds apps fast, but security gaps can expose user data. The 10 most common issues in Lovable apps and how to fix each one before launch.
11 min read
research
We Scanned 100 AI-Built Apps. Here's What They All Missed.
We analyzed 100 repos built with Cursor, Lovable, Bolt, and v0. The pattern is consistent: AI tools build fast but leave critical gaps. Here's the data.
8 min read
tools
v0 to Production: What Vercel's AI Misses
v0 generates beautiful UI with Tailwind and shadcn, but UI isn't an app. What v0 gets right, what it misses, and how to take its output to production.
9 min read
guides
The Vibe Coder's Production Launch Checklist (2026)
A tool-agnostic pre-launch checklist for vibe-coded apps: security, testing, error handling, deploy config, performance, SEO, and monitoring. No fluff.
7 min read
industry
How Much Code Is AI-Generated in 2026? Every Major Figure, Explained
Developers say AI agents fully write about 47% of their code; Google says 75% of its new code. What each figure measures, plus quality and security data.
10 min read
guides
How to Ship Your AI-Built App: A Production Readiness Guide
You built something with Cursor, Bolt, or Lovable. Shipping it is a different challenge. A practical guide to closing the last-mile gap.
10 min read
guides
The AI Code Finishing Checklist: 47 Things to Check Before You Ship
A prioritized checklist for taking your AI-built app from 'it works' to 'it's ready.' P0 items block shipping, P1 within a week, P2 before scaling.
14 min read
guides
Vibe Coding Is Real. Here Is How to Do It Responsibly.
Vibe coding (building software by feel with AI tools) is how millions of developers work now. Here's how to keep the speed without shipping the risk.
14 min read
tools
Cursor vs Lovable vs Bolt: What Each Tool Gets Right (and What They All Skip)
An honest comparison of Cursor, Lovable, and Bolt: what each does well, the gaps all three leave, and how to close them before you ship.
14 min read
deployment
How to Deploy Your Next.js App to Production (The Complete Guide)
The gap between npm run dev and a production deploy is where most AI-built apps die. A step-by-step guide to getting your Next.js app live and reliable.
14 min read
testing
Testing AI-Generated Code: A Minimum Viable Test Strategy
AI tools generate code but never tests. A minimum-effort testing strategy that catches the bugs that matter before your users find them.
12 min read
security
Why 45% of AI-Generated Code Fails Security Tests (And How to Fix It)
The five most common security failures in AI-generated code, and how to catch them before your users do.
13 min read
security
The Complete Guide to Supabase Row Level Security for AI-Built Apps
Supabase is the default backend for AI-built apps. Without proper RLS policies, user data is one API call from exposure. Here's how to lock it down.
16 min read
guides
Error Handling in AI-Built Apps: The Patterns AI Never Generates
AI tools generate the happy path, but production is the unhappy path. Six error handling patterns AI keeps skipping, and the code to fix each one.
17 min read
tools
What a FinishKit Scan Actually Finds (Real Results from 5 AI-Built Apps)
We ran FinishKit on five real AI-built apps, from a Cursor SaaS to a Lovable marketplace to a Bolt landing page. Here's every finding, ranked by severity.
15 min read