@mattpocockuk: Experimenting with a rare addition to my global CLAUDE.md: "Always talk in ASD-STE100 Simplified Technical English. Alw…
Summary
Matt Pocock experiments with adding a global CLAUDE.md instruction to use ASD-STE100 Simplified Technical English and read CONTEXT.md files for ubiquitous language.
View Cached Full Text
Cached at: 08/05/26, 12:17 AM
Experimenting with a rare addition to my global CLAUDE.md:
“Always talk in ASD-STE100 Simplified Technical English. Always read CONTEXT.md files, and use their ubiquitous language.”
Similar Articles
@mattpocockuk: Debating renaming CONTEXT.md to GLOSSARY.md Over time CONTEXT.md has shrunk to literally just a glossary, so this makes…
This describes a set of agent skills for real engineering with coding agents like Claude Code and Codex, aimed at fixing common issues in AI-assisted development.
multica-ai/andrej-karpathy-skills
A single CLAUDE.md file that implements four principles to improve Claude Code's coding behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
@jbarbier: For those starting with AI coding, I just shared my CLAUDE.md (also works with Gemini and Codex BTW - see how-to). Sinc…
Developer Julien Barbier shares his CLAUDE.md configuration file for AI coding agents, which improves efficiency by providing explicit instructions for Claude, Gemini, and Codex. The file is customizable and includes setup instructions for multiple tools.
@PawelHuryn: 187,000 people have starred one CLAUDE.md file on GitHub. A CLAUDE.md loads on every turn. Every line is rent you pay o…
A Twitter thread by Paweł Huryn shares insights on writing effective CLAUDE.md files for Claude Code, emphasizing project-specific rules and avoiding unnecessary bloat.
@dair_ai: If you maintain an AGENTS.md or a CLAUDE.md, this is worth a read. (bookmark it) 288 gold-test evaluated runs across Cl…
This paper presents a controlled ablation study across Claude Code and Codex, 17 real tasks, and 288 runs, finding that context files like AGENTS.md/CLAUDE.md do not measurably improve correctness; agents fail on implementation skill, not missing repository knowledge.