Show HN: Distill anyone's writing style into a skill

Category: ai-ml

Tags: stylometry, prompt-engineering, agent-skills, writing-style, text-analysis

Score: 6.0/10 (Innovation: 6, Technical: 5, Documentation: 8, Utility: 5)

Mimic distills an author's writing style into an agent skill by combining stylometric fingerprinting (L1) with rhetorical move extraction (L2), producing a system prompt that mimics the voice without fine-tuning. It includes a benchmark showing blind LLM judges prefer the skill's output over plain 'write as X' prompts, though surface stylometry can't distinguish them. The project is interesting for applying measurable structural statistics to prompt engineering.

Target audience: AI engineers, prompt engineers, writers, and agent developers interested in style transfer and persona simulation

Repository: https://github.com/xatuke/mimic · Python · MIT · 4 stars

View on Hacker News