Point your coding agents at Clean Code instead of rule lists — it works
Al_Grigor · x · 2026-10-22
Instead of enumerating rules, tell your coding agents to follow a classic book — e.g. "follow the principles from Clean Code by Robert Martin." The author tested this on a large codebase built entirely with AI since 2023 (first copy-pasting from ChatGPT, then fully agent-driven), which had degraded into slop. Asking the agent to refactor against the book's principles took just two banked Codex resets and made the code readable again.
He also cites Hugo Bowne-Anderson's writing variant: put Strunk & White's Elements of Style rules ("omit needless words," "use the active voice") into AGENTS.md, run three parallel verification agents over the draft, then a final editor agent — because agents don't always follow the rules.
More from coding & agent
- Grok Bot Lands in X DMs: An Autonomous Assistant That Connects to Your Apps — mark_k · 2026-10-22
- 5 ways AI makes engineering leaders more effective — beyond writing code — rseroter · 2026-10-22
- ToM2Vec: continuous theory-of-mind vectors help LLM agents coordinate with unseen partners — EliasEskin · 2026-10-22
- LiDAR scan + LLM workflow produces dimensionally accurate 3D room models — OpenAIDevs · 2026-10-22
- Working with agents: give them principles, not rules — _AustinCalvert_ · 2026-10-22
- Quant-Arena: 9 LLM agents trade real A-shares, only 4 profit and none beat equal-weight — mhdfaran · 2026-10-22