# How to Vibe Code Your Own Keyword Chef (and Stop Paying for It) > Rank Insanely Fast for Keywords Your Competition Can’t Find - Site: https://keywordchef.com - Category: SEO & Keyword Research SaaS - Platforms: Web app - Verdict: **Solid side project** (68/100 vibecodeable) - Estimated effort: 2-3 weeks part-time ## Verdict You can build a fully functional personal clone of Keyword Chef, but handling proxy rotation and SERP anti-bot blocks will require patience. The dashboard, wildcard permutation logic, credit balance ledger, and keyword clustering UI are straightforward to generate using an AI coding agent. The real engineering friction lies in the data pipeline: hammering Google Autocomplete and scraping live search result pages in bulk without getting your IP banned or hitting persistent CAPTCHAs requires setting up a reliable proxy rotation layer and fallback parsing logic. ### What you can't replicate - The community trust and brand equity built over years among niche site builders - The exact historical edge cases baked into the founder's heuristic filter rules ## What it does A specialized niche-research and keyword-discovery tool that automatically filters junk keywords, evaluates live SERP scores to highlight weak competitor spots, and features smart wildcard expansions for content publishers. ### Core features - Wildcard search matrix generation (e.g. 'best * for chefs') - Google Autocomplete data harvesting - Bulk live SERP scraping and first-page result parsing - SERP Score calculation (identifying forums and user-generated content) - Automatic junk keyword filtering and intent classification - Keyword clustering and geo-targeted search - Credit ledger and pay-as-you-go balance management - Public report sharing links ## The business ### Pricing - Free Trial: Free — Introductory credit bundle to test searches and SERP analysis. - Credit Packs: $20+ — Pay-as-you-go non-expiring credit system for wildcard searches and bulk SERP checks. Founded 2021. Team size: 1 (Solo-founded). ## The hard parts - Bypassing Google and search engine anti-scraping blocks, bot detections, and rate limits at scale - Managing rotating proxy pools, user-agent spoofing, and handling CAPTCHAs reliably - Asynchronous high-throughput harvesting of Google Autocomplete permutation trees - Tuning heuristic rules for the SERP Score algorithm to accurately spot forum domains and weak sites ## How to vibe code Keyword Chef ### Prerequisites - Node.js (free): Runtime environment for building and running the full-stack web application. - GitHub (free): Version control and deployment integration. - Scraping Proxy Provider (~$15/mo): Required for fetching live search results and autocomplete data without getting blocked by Google. ### Recommended AI tools - Claude Code: Best-in-class agentic coding tool for scaffolding multi-file apps, handling backend search scripts, and iterating on UI components. - Cursor: Ideal for fine-tuning dashboard views, filter tables, and reviewing code diffs interactively. ### Stack - Frontend: Next.js with Tailwind CSS and shadcn/ui components - Backend: Next.js Server Actions and API routes handling asynchronous scraping jobs - Database: Turso (SQLite at the edge for storing reports, search history, and user credits) - Auth: better-auth (Self-hosted TypeScript auth with email/password for single-user personal access) - Payments: Skipped (personal-use build) - Other: Cheerio or Firecrawl for parsing scraped HTML SERP pages, Axios with proxy agents for Google Autocomplete harvesting ### Hosting - Cloudflare (Hosting the Next.js frontend and edge API routes, paired with Turso for edge database storage.): $0-5/mo ### Build guide 1. **Scaffold Project & Database Schema** — Initialize the Next.js application, configure Tailwind CSS, set up better-auth, and connect Turso SQLite to manage searches, reports, and credit balances. ``` Create a new Next.js project using TypeScript and Tailwind CSS. Configure better-auth with email/password authentication backed by a Turso SQLite database. Set up database tables for `users`, `credits`, `reports` (id, query, wildcard_pattern, created_at, status), and `keywords` (id, report_id, phrase, volume, serp_score, has_forums, cpc). Establish clean folder structures under `app/` and include basic layout styling matching a clean, dark/light SaaS dashboard. ``` 2. **Build Wildcard Matrix & Autocomplete Engine** — Implement the core keyword expansion logic that takes wildcard queries (e.g. 'best * for chefs') and queries Google Autocomplete permutations asynchronously. ``` Build a backend utility in Next.js that accepts a wildcard query pattern containing an asterisk (e.g., 'best * for chefs'). Generate an alphabetized matrix of variations (a-z) and query the public Google Autocomplete suggest endpoint asynchronously for each variation with randomized user-agents and proxy support. Deduplicate results, clean up junk phrases, and return a consolidated array of candidate keyword suggestions. ``` 3. **Implement Bulk SERP Scraping & SERP Score Algorithm** — Create a background scraping pipeline that fetches first-page Google search results for candidate keywords and calculates the SERP score. ``` Implement a bulk SERP analysis module. For a given list of keywords, trigger an asynchronous scraper using proxy rotation to fetch Google search result pages. Parse the HTML to extract the top 10 ranking domains. Write a heuristic matching algorithm that checks domain lists against known user-generated content and forum patterns (Reddit, Quora, Pinterest, medium, etc.). Calculate a 'SERP Score' representing how many weak/forum spots appear on page one, and persist the parsed results into the `keywords` table. ``` 4. **Build Keyword Filtering & Clustering Dashboard** — Develop the primary user interface featuring interactive filter controls, SERP score badges, keyword intent filters, and clustering views. ``` Build a responsive keyword report dashboard page in Next.js. Display results in a filterable, sortable data table showing keyword phrase, estimated search volume, SERP score, and a badge highlighting if forums rank on page one. Add interactive filter controls to filter out low-intent phrases, sort by SERP score, and group related queries into automatic keyword clusters. Include a modal view to inspect the live first-page SERP breakdown for any specific keyword. ``` 5. **Add Credit Ledger & Report Export** — Implement credit consumption tracking for searches and provide export functionality for saved keyword reports. ``` Implement a credit ledger system where each wildcard search or bulk SERP check deducts a defined amount from the user's balance. Display remaining credits in the header. Add an export button to download keyword reports as a CSV file, and implement public report sharing links where generated reports can be viewed via a unique read-only URL. ``` ### Cost vs paying **Starting costs (one-time):** - Scraping Proxy Pool Starter: $15 - Domain name (optional): $12 - Total: ~$27 one-time **Ongoing costs (monthly):** - Cloudflare & Turso edge hosting: $0 - Proxy bandwidth for SERP scraping: ~$15/mo - Total: ~$15/mo - Paying for the SaaS instead: $20 credit packs - Build time: 25-35 hours - AI tool credits: $20 (Claude Pro) - Break-even: N/A (Personal build for custom workflow) ## Sources - [Keyword Chef Official Website](https://keywordchef.com) - [LowFruits vs Keyword Chef 2026](https://vertexaisearch.cloud.google.com/grounding-api-redirect/AUZIYQE9A1zGnQOTGl-p84-qJcwYcBDWOtZQxWtUwNdYCAI296pTyKZxV7lThRIfSqUAn5Cjcc-UNlKmenWmVd4_8345hd2dpDvVwHlXVzEhr1XUuxztpCbl6e3GZ55giwyrN8ClTl0RMhNnVuSsMr1yBihHSSc=)