Asking Codex to Write a HUMANS.md So AI and Humans Understand Each Other
BLUECOW009 · x · 2026-09-21
@xikhar shared a playful inversion of the CLAUDE.md/AGENTS.md convention: asking Codex to write a 'HUMANS.md' — a doc where the AI describes humans — so the two sides can understand each other better.
It flips the usual 'humans write instructions for AI' pattern on its head, and doubles as a lighthearted take on human-AI collaboration norms.
More from coding & agent
- Dev reveals ghost agent tool unreleased because the UI is terrible — BLUECOW009 · 2026-09-21
- PM builds a real SaaS in 24 hours with AI agents, sharing his full zero-cost stack — PawelHuryn · 2026-09-21
- NPC-Forge: Deterministic CPU-Only Agents That Run on a Raspberry Pi — gioscarab · 2026-09-21
- Prompt cache restore works, but reusing KDA/Mamba states hits illegal memory access — TheZachMueller · 2026-09-21
- Engineer: Big Company Ships Everything Via Claude Code, Nobody Reads It Anymore — Simon Willison · 2026-09-21
- Why this dev hasn't shipped his agent: heavy tokens, slow, CPU-hungry — BLUECOW009 · 2026-09-21