A prompt recipe for making Claude Code sound less like AI-generated slop
wzenus · x · 2026-08-04
A user shared an instruction set for making Claude Code sound more natural and less like AI-generated slop. The prompt pushes the model toward standard terminology, direct phrasing, neutral table labels, no metaphors, no “X not Y” constructions, and no anthropomorphic language.
The accompanying comment says this kind of system prompt is now worth paying extra for, because Claude-style writing has become recognizable and often feels like AI slop unless it is tightly constrained. The post is essentially a practical prompt recipe for cleaning up coding-agent output.
Related event: Prompt Rules Shared to Remove Claude Code's AI Tone(3 posts)→
More from coding & agent
- Stanford’s CS239A course on self-improving AI agents is now on YouTube — mariofilhoml · 2026-08-04
- Cline says open-weight models work better when the harness lets them verify more — cpaik · 2026-08-04
- WorkOS schedules Agent Night in San Francisco with context-graph talks and live demos — davidcrawshaw · 2026-08-04
- A curated reading list for DeltaNet, FlashKDA, vLLM serving and MoE — austinvhuang · 2026-08-04
- Yann LeCun says strong code-generation systems go beyond plain autoregressive LLMs — ylecun · 2026-08-04
- A multi-agent workflow: Fable 5 plans, Codex executes, Fable verifies — dotey · 2026-08-04