Taming AI Jargon: A Prompt Guide to Make Claude Code Speak Clearly
wzenus · x · 2026-08-04
AI coding assistants often output obscure metaphors or invented terms (e.g., calling a checkpoint an "archive" or a base model a "bloodline"). A developer shared a prompt guide to constrain Claude Code to use plain language in scientific and engineering tasks.
The instructions require the model to use standard terminology, avoid metaphors and anthropomorphism, and keep neutral headings. Examples:
- Terminology fixes: Change "super not super long" to "whether the length limit is exceeded"; change "farming rewards" to "reward is optimized but eval metrics did not improve".
- Removing metaphors: Change "bloodline" to "base model source"; change "error margin still covers the baseline" to "confidence interval overlaps with the baseline".
More from coding & agent
- Can AI Coding Agents Crack Nvidia's CUDA Moat? — bingxu_ · 2026-08-04
- HeyGen LiveAvatar Agent Builds $3M Pipeline During 8-Week Employee Leave — HeyGen · 2026-08-04
- Agent Benchmark Reflections: Scores Are Deceptive, Open-source Trajectories Needed — Shahules786 · 2026-08-04
- Agent Benchmark Flaws: Over-specified Verifiers Penalize Semantically Correct Actions — Shahules786 · 2026-08-04
- ITSMBench: Open-Sourcing a Benchmark for Enterprise AI Agents — Shahules786 · 2026-08-04
- Protecting AI Attention: The Essence of Inference Efficiency — DanWahlin · 2026-08-04