Experiment: attaching an unslop style skill to a model also rewrites its chain of thought
eliebakouch · x · 2026-08-24
The author ran an experiment: after attaching poteto's unslop skill (which strips AI-slop writing patterns) to a model, the skill changed not just the output style but the model's chain of thought — even explicitly instructing the model not to apply the rules during reasoning made no difference.
Questions were generated by Claude; the author also noted DeepSeek V4 Pro almost never uses em dashes, unlike many other models. It suggests system-level skills have deeper side effects than expected.
Related event: Style Skills Quietly Rewrite Model Chain-of-Thought(2 posts)→
More from coding & agent
- Structured conflict beats scaling agents in code review: new study — omarsar0 · 2026-08-24
- Speeding Up Iteration with Opus-5 Cuts Experiment Decision Time by 50% — morgymcg · 2026-08-24
- Returning to Gemini: which models are good for software dev? — biloo0asks · 2026-08-24
- The Tradeoff of AI Coding Agents: Losing Manual Edge — bendee983 · 2026-08-24
- gcontext introduces allowlist to control MCP tool exposure — javigarcia11093 · 2026-08-24
- Best local coding model for 6GB VRAM: Balancing Qwen performance and response times — BrianScottGregory · 2026-08-24