The Spec Structure v0, Lovable, and Bolt Actually Want
One-line prompts force app builders to guess. The six-part spec skeleton that gets v0, Lovable, Bolt, and Replit Agent building the app you actually meant.
Read
Blog
Practical writing on prompting AI coding tools. No hype, worked examples, revised as the tools change.
Cursor Prompts vs Claude Code Prompts: Why the Same Prompt Fails in Different Tools
The same JWT auth task written for Cursor Agent mode and for Claude Code, side by side, with every difference explained and when to use each tool.
Read
GitHub Copilot Prompts That Actually Work: Inline Chat, Edits, and @workspace
Copilot is file-scoped, not repo-aware. Prompt patterns for inline chat, Copilot Edits, and @workspace, with full examples and a fix for the biggest mistake.
Read
How to Prompt Agentic Coding Tools: Why 100 Words Beat 500
Why short prompts with hard constraints and a verification step beat 500-word mega-prompts in Cursor, Claude Code, and Copilot, with a full rewrite example.
Read
Landing page prompts for AI app builders: v0, Lovable, Bolt, and Replit Agent
A working library of landing page design prompts for v0, Lovable, Bolt, and Replit Agent — full pages and individual sections, each with a customizer for style, palette, type, and motion.
Read
Why Most AI Prompt Libraries Don't Help You Ship
Most prompt libraries are scraped, undated, and untested. The failure modes that make them useless, and what to look for before you trust one.
Read