A simple prompt trick: tell the model to shrink any output 75% without losing function
kuza55 · x · 2026-09-06
kuza55 shares a simple prompting trick: take any model output and instruct it to make it 75% smaller without losing functionality. It reliably produces tighter, more focused results—useful for trimming verbose answers or bloated code.
More from coding & agent
- Dev builds an image watermarking web app with Grok Build Mode — nima_owji · 2026-09-06
- huashu-mac-use open-sourced: give any agent near-GPT-6 computer-use skills — AlchainHust · 2026-09-06
- Multitouch jelly baby: Three.js demo adds grabbable mobile physics — DavidWells · 2026-09-06
- Mac Studio user vibe-codes a local AI cluster monitor app in hours, prompt included — SquirrelEStuff · 2026-09-06
- Microsoft's MarkItDown tops GitHub with 100k+ stars, turning messy docs into LLM-ready Markdown — mdancho84 · 2026-09-06
- Dev builds keyframe rigging editor that feeds JSON edits back to LLMs — tomjohndesign · 2026-09-06