Describe an app in a browser chat and get a working, hosted, full-stack build — the course's default MVP factory.
Pick this whenYou want the best-looking first draft fastest and you'll graduate the code out later — vs Bolt (weaker), Replit (more IDE), or Base44 (faster but less polished).
What makes it different
Won the bake-off on design — 'the best-looking output' across five platforms tested live in BC4
One-click GitHub sync, so the project can be cloned and taken over by a coding agent
Auth, secrets prompts and a code view are built in — it scaffolds a real app, not a mockup
What it does
Scaffolds a full app from a prompt: front end, routing, auth with email validation, database wiring
Pushes to a GitHub repo you own, so nothing is trapped in the platform
Prompts for the API keys a feature needs (OpenAI key for the AI project generator)
Shows the generated code, so you can learn from it or hand it to Claude Code
What it does not do
Common wrong expectationsDoesn't produce a distinctive look by default — everything ships as the same purple gradient unless you name Tailwind/shadcn or feed it a reference screenshot
Doesn't stay cheap — 'a token eater'; free credits ran out inside one session
Doesn't wire cleanly to Supabase anymore — it steers you to Lovable Cloud instead
Where it sits in a stack
The first 20% of a build. Take the MVP to the point where the idea is proven, then clone the repo and continue in Claude Code or Cursor where iteration is cheaper.
ShelfVibe-coding platforms
Our status · not-usingThe KB is static HTML by design; no app to vibe-code yet.
For your projectsThe realistic candidate if the KB ever needs an interactive front end (search UI, review queue) beyond what build_site.py can emit — but the graduation-to-Claude-Code pattern matters more than the platform.
claude-code: the documented graduation path — clone the Lovable repo and take over
| Session | Coverage | Moment | What happened |
|---|
| Basecamp 2: Vibe Coding | demonstrated | 1:08:32 | Built 'Task AI' end-to-end: scaffold, auth with email validation, secrets prompts, code view, GitHub connect, credit billing view; declined-then-adopted Lovable Cloud |
| Basecamp 4: Introduction to n8n — II | demonstrated | 2:17:35 | Bake-off winner on design; 'a token eater' — free credits nearly exhausted; repo pushed to GitHub and shared |
| Office Hours VI: Paul's SEO/AEO/GEO Question - Moving a Lovable Site into GitHub and Claude Code, the GEO Skill, llms.txt, and CLIs as the Glue | demonstrated | | Mock client site; pushing code out is deliberately hard |
| Session 3: Building Elegant Landing Pages | demonstrated | 1:48:01 | Code-zip download demoed for migration to Codex; endorsed for beginners/prototyping with one-click hosting |
| Session 4: How to Build Micro Prototypes | demonstrated | 0:28:30 | Directory app one-shot from attached CSV, plan mode, AI-assistant mode via Lovable Cloud (model/db/secrets invisible), publish flow, code-zip download, GitHub integration; credit economics dissected |
| Session 16: GitHub 101 — Fundamentals & Essentials | demonstrated | 1:14:55 | Built the Sherlock escape-room app ('The Great Escape Society' / 'Baker Street Disappearance'), connected to GitHub via OAuth, auto-committed 12 times; also generated the README on the dev branch |
| Day 12: MCP Ultra-Hard Mode — Editing the Config Safely, Building Your Own MCP Server in n8n, Opal, Cross-LLM Memory, and an Early Look at Claude Cowork | demonstrated | | Chatbot front end on an n8n webhook |
| Day 13: Front End Meets Backend — Lovable to n8n by Webhook and .env, the Lovable<->n8n MCP Connector, Respond to Webhook, and GitHub for Teams | demonstrated | | .env file, n8n connector, generated expense-tracker UI |
| Day 2: ACR and DRA Thinking Frameworks, Connectors and Scheduled Tasks, the Image-to-Video Ad Pipeline, MoSCoW + 5 Whys, and Building LearnVault Back-End-First in Lovable | demonstrated | | LearnVault build |
| Bonus: Launch a Digital Product With AI in Two Days — Quick-Win Products, Order Bumps, Seven Myths, and the Five-Step AI Build | demonstrated | | Sales copy to page mockup in ~1 minute |
| Vibe Coding for Non-Coders: Pain Point to MoSCoW to Stack, 'Use AI to Build AI' with Grok, and an Audio Note-Taker Built Live on Lovable + Supabase + OpenAI | demonstrated | | Front end; Manage Supabase; Try to Fix; Publish |
| Learning Topics: Sessions 1–5 — Foundations to AI Director | explained | | The live CRM case study: planning mode, Lovable Cloud backend, per-user auth, publish failure debugged, security scan |
| Workbooks 1–5: Guided Builds with Completion Proof | explained | | Workbook 2's guided builder option (vs Bolt.new as the fast option) |
| Building a Web App with Replit: Vibe Coding Explained, the Tool Landscape, and a Live Build of a Business Site with an AI Sales Chatbot | explained | | |
| Office Hours I: Vetting an Idea, Niching Against the Frontier Labs, and Paul's Aircraft-Parts App (PWA, and 'What If I Get Hit by a Bus') | mentioned | | Paul's build tool for the parts app |
| Office Hours V: Paul's Compliance Scanner Meets Firecrawl, Painkillers vs Vitamins, What a Real Second Brain Costs, and Why Selling Got Harder | mentioned | | Paul's front end for the scanner |
| Office Hours XII: Building Paul's Second Brain — the Full Project-Management System, Live | mentioned | 0:09 | The build target whose credits started the whole question |
| Session 14: Claude Code — Advanced Workflows | mentioned | 2:03:54 | The comparison baseline the shadcn skill beats on UI cleanliness |
| Session 15: Loop, Harness & Graph Engineering (posted as 'Securing Vibe-Coded Apps') | mentioned | 0:15:44 | 'Peanuts' next to Claude Code — the beginner rung, with Bolt |
| Session 17: Facebook Ads Competitor Research & Replication | mentioned | 0:50:47 | Named as the UI wrapper if the workflow were productized fully |
| Build Your 2026 Marketing and Distribution Engine: the GTM Loop, Painful-Reachable Markets, the Evidence Ladder, the Six-Part Offer, Create vs Capture Demand, Channel Mapping, and Three Playbooks | mentioned | 0:30 | Shareable build URLs |
| Founder Stack Ep03 with Shai Krishna: Why Wrappers Make AI Slop - CLAUDE.md as the Brain and DESIGN.md as the Art Direction, Founder-Led UGC with AI Avatars, Trial Reels, and Email Journeys | mentioned | 0:15:00 | |
| Bonus + Demo Day: Self-Hosting n8n with Docker (Cloud vs Local Decision, the n8n_data Volume, Credentials, ngrok) and How the Hackathon Was Judged | mentioned | | TripTally front end (Bolt evaluated first) |
| Day 8: AI Workflows — Chaining Tools, the IDEA(L) Analysis Framework, Claude Skills vs MCP vs Projects, and the Reel-Cloning Pipeline | mentioned | | App builder pick |
| Day 14: Graduation — Skills, Systems, Dollars: the Niche Framework, the First-Dollar Roadmap, Pricing, Moats, and What Happens Next | mentioned | | Front ends for productized systems |
| Beginner Glossary (30 terms) + I'm Stuck Guides (23) | mentioned | | The Lovable/Bolt-vs-Claude-Code guide: continue with the tool that fits; export a working version before major changes |
| Practice Lab: Six Confidence-Building Projects | mentioned | | Named among the builder options for projects 1–2 (with Bolt, Replit, Claude Artifacts) |
| The Five-Level AI Generalist Roadmap (+ Operating System & Monetization) | mentioned | | L5 non-technical track with Bolt, v0, Replit; also the micro-SaaS prompt-first stack |
| Chapter 1: The Mindset — Figma Is a Sandbox, and Your Brain Is the Obstacle | mentioned | | AI builders named and explicitly placed out of course scope |
| AI Sprint: AGI Readiness - Day 1 (Three Claims and Three Dials, the Agent Harness and Autonomy Ladder, the Failure Mode of Abundance, and the Six-Layer Readiness Stack) | mentioned | | Delegation/orchestration layer |
| AI Sprint: Claude Code with Antigravity — Day 2 (CLAUDE.md, the 50% Context Rule, and the Sketch-to-App Build) | mentioned | | The named inspiration, treated respectfully: 'don't expect it to work exactly like Lovable' |
| AI Sprint: The Hermes Agent — Day 2 (Loop Engineering: the Scheduling Loop, the Delegation Loop, and the Skill Auditor) | mentioned | | Setup for the vibe-coding awareness story — the room outside doesn't know the words |
| AI Masterclass: Exploring Antigravity 2 — Four Products, Four Slash Commands, and Two Live Builds | mentioned | | His 'only good for mockups' claim — flagged as partisan; the corpus's Lovable projects say otherwise |
| AI Sprint: The Open Source Playbook — Day 2 (Weights 101, OpenCode, and the Subsidy Math) | mentioned | | Named as one of Anthropic's biggest B2B customers — the subsidy math's anchor |
| AI Sprint: Windsurf & Devin — Day 1 (Vibe to Agentic: GitHub as Pizza, the Four Spec Concerns, and the IDE Lineage) | mentioned | | Named in the vibe-coding conversation tier with Bolt and Emergent — descriptive, not dismissive |
Lovable