I built a skill that makes LLMs stop making mistakes

Over the past few weeks, I noticed a recurring pattern: people around me were manually appending **“make no mistakes”** to the end of their Cursor prompts. It was one of those small workflow habits that everyone had adopted individually, simple, but repetitive. Typing the same instruction again and again felt like unnecessary friction in a toolchain that’s supposed to optimize productivity. So I built **make-no-mistakes**. The project automates this step by ensuring that the instruction is consistently applied without requiring manual input each time. It removes the repetitive overhead and lets developers focus on what actually matters: writing and iterating on code. In a world where we automate CI pipelines, deployments, testing, and infrastructure, manually typing the same prompt suffix over and over again feels oddly out of place. **make-no-mistakes** fixes that. Gstack is in the past, I urge everyone to move to Mstack. You can explore the project here: [https://github.com/thesysdev/make-no-mistakes](https://github.com/thesysdev/make-no-mistakes) It’s 2026. Stop. Shipping. Slop.