# How to Vibe Code Your Own Scalenut (and Stop Paying for It) > AI-powered SEO and Generative Engine Optimization (GEO) platform - Site: https://scalenut.com - Category: Marketing & SEO SaaS - Verdict: **Serious undertaking** (48/100 vibecodeable) - Estimated effort: 3-4 weeks of serious engineering part-time ## Verdict Build a personal subset for keyword clustering and LLM article drafting, but keep paying if you need reliable multi-engine AI visibility tracking. Cloning a personal version of Scalenut's core article writer and keyword clustering engine is feasible with a modern TypeScript stack, but building reliable AI visibility scrapers that query ChatGPT, Perplexity, and Google AI Overviews weekly without getting blocked by anti-bot measures is an ongoing maintenance nightmare. You will spend more time fixing scraping selectors and managing API rate limits than doing actual SEO work. ### What you can't replicate - Historical visibility tracking data across millions of tracked prompts - Direct enterprise partnerships and built-in backlinks marketplace liquidity - Managed senior strategist human-in-the-loop services ## What it does An end-to-end GEO workspace that plans content strategy, monitors brand citations across generative engines like ChatGPT and Perplexity, and generates long-form SEO articles with custom brand voices. ### Core features - AI Visibility Tracker across ChatGPT, Perplexity, and Google AI Overviews - Keyword cluster generator and topical authority mapper - Long-form SEO/GEO article writer ('Cruise Mode') with NLP term scoring - Automated content auditor and page optimizer - CMS auto-publishing integrations (WordPress, Shopify) - Internal link mapping engine ## The business ### Pricing - Starter: $59/mo — For founders and small teams getting started with AI visibility - Plus: $89/mo — For growing teams that want actionable insights and optimization - Professional: $199/mo — For advanced teams building AI visibility as a core growth channel ### Funding $3.5M - $6.5M raised. - Seed Investors: Decibel Founded 2020. Team size: 28-60. ## The hard parts - Multi-engine AI visibility scraping that evades rate limits and dynamic DOM changes across LLM chat interfaces - Real-time SERP parsing and NLP entity extraction to compute live content optimization scores - Multi-step asynchronous LLM pipelines ('Cruise Mode') that chain outline generation, factual research, and brand voice adherence without failing ## How to vibe code Scalenut ### Prerequisites - Node.js (free): Runtime environment for the Next.js full-stack application - GitHub (free): Source code repository and CI/CD integration - Cursor (Free tier / $20/mo Pro): AI-native code editor for building and iterating on complex frontend and backend workflows ### Recommended AI tools - Cursor: Best code editor for multi-file component creation and iterative refactoring of AI pipelines - Claude Code: Autonomous terminal agent for scaffolding complex database schemas, API routes, and background queue workers ### Stack - Frontend: Next.js with Tailwind CSS and shadcn/ui - Backend: Next.js API routes with Vercel AI SDK - Database: Turso (SQLite at the edge) - Auth: better-auth - Payments: None (personal use clone) - Other: Inngest for background article generation queues, Exa API for semantic web search and research gathering ### Hosting - Vercel (Hosting the Next.js frontend and serverless API endpoints): $0-20/mo - Turso (Serverless SQLite database for workspaces, keywords, and generated articles): $0/mo ### Build guide 1. **Scaffold Project and Authentication** — Initialize a Next.js project with Tailwind CSS, shadcn/ui, and configure better-auth with Turso as the database backend. ``` Create a new Next.js project using Next.js 16, TypeScript, Tailwind CSS, and App Router. Set up better-auth with email/password authentication connected to a Turso SQLite database using the `@libsql/client` driver. Create the core database schema tables for users, workspaces, projects, and articles with proper foreign key relations and indexes. Implement a clean, responsive dashboard layout with a sidebar navigation mimicking an SEO SaaS workspace selector. ``` 2. **Keyword Cluster & Topic Planner Engine** — Build an AI-powered keyword planner that takes a seed keyword, calls an LLM via Vercel AI SDK with Exa search grounding, and structures hierarchical topical clusters. ``` Build a keyword planning module in Next.js using the Vercel AI SDK and Anthropic API. Create an input form where users enter a seed keyword and target country. Write a server action that queries the Exa search API for related queries, passes the results to Claude with a structured JSON output prompt, and returns a categorized topical cluster list with search intent, difficulty estimates, and parent-child topic hierarchies. Store the generated clusters in Turso and display them in a responsive data table with filtering and export capabilities. ``` 3. **Cruise Mode Long-Form Article Generator** — Implement a multi-step asynchronous article generation pipeline using Inngest to research, outline, draft, and optimize long-form content. ``` Implement an asynchronous long-form article generation pipeline using Inngest and the Vercel AI SDK. The workflow must execute four sequential steps: 1) Generate an optimized outline based on top search competitor headings retrieved via Exa; 2) Gather factual research and talking points; 3) Draft each section iteratively with custom brand tone instructions; 4) Compute an NLP term coverage score against target keywords. Save intermediate states to Turso and provide a real-time progress indicator on the frontend polling the job status. ``` 4. **Content Optimizer and NLP Scoring Dashboard** — Create a live content editor interface with a real-time NLP term checklist and content scoring algorithm. ``` Build a rich-text content editor interface in Next.js using a modular block or markdown editor component. Alongside the editor, create a live sidebar panel that analyzes the current document text against a list of required NLP keywords and semantic entities. Calculate a live 'GEO Score' from 0-100 based on keyword density, heading structure, and word count. Highlight missing terms dynamically and provide a button that uses an AI endpoint to suggest contextual sentences incorporating missing terms. ``` 5. **AI Visibility Tracker Dashboard** — Build a prompt tracking module that queries LLM endpoints weekly to check for brand citations and presence. ``` Build an AI Visibility Tracker dashboard that allows users to input target brand names, target domains, and a list of prompts (e.g. 'Best CRM software'). Set up a Cron endpoint protected by a secret token that invokes OpenAI and Perplexity APIs to evaluate whether the target brand is mentioned in the generated answer. Store historical check results in Turso with timestamps and display trend graphs using a chart library like Recharts showing brand citation share over time. ``` ### Cost vs paying **Starting costs (one-time):** - Cursor Pro subscription: $20 - Anthropic/OpenAI API starting credits: $25 - Total: ~$45 one-time **Ongoing costs (monthly):** - Vercel Hobby/Pro hosting: $0 - $20/mo - LLM API usage (Claude/OpenAI/Exa): ~$15/mo - Total: ~$15-35/mo - Paying for the SaaS instead: $59/mo - $199/mo - Build time: 40-60 hours - AI tool credits: $20 (Cursor Pro) - Break-even: 1 month vs Professional tier ## Sources - [Scalenut Official Website](https://scalenut.com) - [Scalenut Pricing & Features](https://scalenut.com/pricing)