Programming · Blog · 🌐 International · via agentusecases.ai
A requirements doc shelved for a year becomes a live website in one day with Claude Code
A tech consultant pulled out the spec for an aggregator of his own posts and shipped the site in a single day with Claude Code.
- Profession
- Developer, Consultant
- Tools
- Claude Code, Claude, Netlify
- Source
- dev.to ↗
- Country
- 🌐 International
- Found via
- agentusecases.ai (Alex Finn) ↗
- Report date
- July 8, 2025
- Verified on
- October 30, 2026
See more: AI for developers · AI for consultants
What they did
Justin Wheeler was writing in four different places (his employer's blog, Dev.to, AWS Community Builders and RedMonk) and wanted to bring it all together on one site, with his own URLs, filters and search. The eight-page spec had been ready since July 2024, but nobody took on the project. He used Claude in the browser, over several rounds, to condense the document into a short, structured prompt, then handed that prompt to Claude Code, which built the site in Vue 3, TypeScript, Tailwind and Vite. Hosting went to Netlify, connected to the GitHub repository.
How it went
According to the author, the site was built and published in one day, with a single list of posts, content driven by a JSON file, custom URLs with redirects, filtering by date and platform, search and dark mode. He says the result exceeded his expectations and that the decisive step was turning the long document into a focused prompt. The code is open on GitHub.
Good to know
A long, scattered requirements document gets in the agent's way: summarize it first into stack, features, data structure and design rules. Review the generated code before putting it live on your own domain.
The prompt
I have this requirements document for a project: [paste document]. Turn it into a prompt for a coding agent with four sections: required stack [technologies], features in priority order, data structure with an example, and design rules. Cut anything that is opinion or an edge case, and list at the end the questions I need to answer before starting.
Replace what is in [brackets]. Prompt written by us from the account, not a literal copy of the source. Do not paste other people’s personal data (privacy laws such as LGPD and GDPR apply).
Summary in our own words. Results as stated in the original account; we have not tested or audited them. Read the source ↗ Case found via agentusecases.ai (Alex Finn); text and prompt rewritten by us. Found a mistake?
Similar cases
01Letting the agent deploy and read production errors on its own
The founder of a deployment platform explains how they designed an MCP server to take the human out of "copy and paste the log".
02Teaching an agent to use your bank's API
A developer created an open source skill so Codex and Claude Code can integrate with the Banco Inter API.
03Leave a coding agent running for 24 hours and learn from the bill
An experiment letting Claude Code work unsupervised for a full day ended with a high bill and a security scare.
04Switching from Cursor to Claude Code for Big Features
A Brazilian dev who works for clients abroad describes how their way of working with agents changed.
05Converting Map Data with AI's Help
Simon Willison used Claude and ChatGPT to convert a shapefile of a park boundary into GeoJSON.
06Livelo engineers report up to 50% higher productivity with AI agents
The Brazilian loyalty program brought Kiro and Amazon Q Developer to its team of about 240 engineers, with metrics and governance from day one.
07A leaner MCP server finds wasted AWS spend in fewer calls
A cloud architect rebuilt a 15-tool MCP server with just 4 tools, and the agent now answers where an AWS account is wasting money in 2 calls instead of 7.
08Launching a habit app solo and doing the marketing with AI
A dev created Kada, a habit app with no account and no ads, and used AI for marketing.