@IndieDevHailey: 82.5k stars, 18 skills, specializing in fixing AI coding derailments! TypeScript guru Matt Pocock (former Vercel engineer, the clearest TS teacher) has open-sourced his real daily workflow with Claude! Four pain points: miscommunication…

X AI KOLs Timeline Tools

Summary

TypeScript guru Matt Pocock open-sourced his real daily workflow skill library for using Claude, containing 18 skills that can be installed with one command via npx skills. It addresses issues in AI programming such as communication misalignment, inconsistent terminology, lack of feedback loops, and code entropy.

82.5k stars, 18 skills, specializing in fixing AI coding derailments! TypeScript guru Matt Pocock (former Vercel engineer, the clearest TS teacher) has open-sourced his real daily workflow with Claude! Four pain points: miscommunication → output not what you want, inconsistent terminology → verbose output like filler, lack of feedback loop → code never works, code entropy → increasingly messy. His skill library precisely tackles each pain point: /grill-me: deeply interrogate requirements, question every idea like an interrogator /tdd: enforce red-green-refactor, write failing tests first then code, rock solid /caveman: slash token usage by 75%, super concise chat /improve-codebase-architecture: regularly maintain codebase, keep it cleaner as you write. Behind it are decades of engineering experience distilled from The Pragmatic Programmer, DDD, and XP. One-line install: npx skills@latest add mattpocock/skills. Pick some skills, run a setup, and you're done. Recommended to try /grill-with-docs first — you'll find AI finally understands what you're saying.
Original Article

Similar Articles

@bkdgiffug: Still frustrated with AI writing code that makes things worse? Former Vercel engineer Matt Pocock has open-sourced his 18 battle-tested playbooks, specially designed to fix all kinds of AI collaboration failures. This guy who makes TypeScript crystal clear has turned his daily skill set into "brake pads" for AI programming — to let…

X AI KOLs Timeline

Former Vercel engineer Matt Pocock has open-sourced his AI coding agent skill set mattpocock/skills, containing 18 practical methods (such as /grill-me, /tdd, /caveman) to help developers better control AI-assisted coding and avoid writing bad code. It can be installed via npx skills or the Claude Code plugin.

@XAMTO_AI: A veteran engineer packed decades of practical engineering experience into this open-source project, which shot to #1 on GitHub trending, amassing 124k stars. The author, a former Vercel engineer who participated in early Next.js development, compiled 16 practical techniques for collaborating with Claude, installable with a single command. The most impressive…

X AI KOLs Timeline

Former Vercel engineer Matt Pocock open-sourced a project called 'skills' that provides 16 practical tips for collaborating with AI coding agents like Claude, including 'Grill Me' and red-green test cycles. Aimed at solving common AI development issues, it has garnered 124k stars.

@Ryrenz: Whoa, someone open-sourced their entire Claude Code configuration honed over six months. This reference repo has racked up over 9700 stars on GitHub — it's a production-grade setup the author refined by running a real TypeScript microservices project for 6 months. It specifically tackles a headache many people have...

X AI KOLs Timeline

An open-source reference library containing production-tested Claude Code infrastructure from 6 months of real-world use, including auto-activating skills, hooks, and modular patterns for TypeScript microservices projects.

@LinearUncle: Software development at this stage has entered the skill era. Matt Pocock skills, superpowers, oh-my-opencode, gstack, lazycodex, etc.—these geniuses' skills are the libraries worth deeply studying, improving, and then using for yourself. Don't…

X AI KOLs Timeline

Matt Pocock has released a set of agent skills designed for coding agents like Claude Code, Codex, etc., including skills like grill-me and grill-with-docs, aimed at improving alignment and reducing verbose output, emphasizing small size, adaptability, and composability.