$ styleprofile score examples/draft.md '<tmp>/demo-writer.json'
[exit 0]
--- stdout
STYLE COMPARISON   1 chunk, 542 words   vs demo-writer.json (55 chunks)

Overall: close   Delta 0.65
  Lower is closer. The reference's own held-out writing at this length (542 words) scores 0.54 typically (95% under
  0.84).

LLM-likeness: a few LLM traits   0.75
  The reference's own writing at this length (542 words) scores 0.16 typically (95% under 0.44); the LLM drafts 7.68.
  Strongest LLM signals: "these" ▲, Em dashes ▲

By area   Delta ÷ the top of the reference's usual range in each area; bars on a log scale
  close up to 1x, somewhat different to 1.5x, clearly different to 2x, very different above
  Voice and stance             1.18x  ████░░░░░░░░░░░░░░░░  somewhat different
  Punctuation                  0.78x  ███░░░░░░░░░░░░░░░░░  close
  Function words               0.76x  ███░░░░░░░░░░░░░░░░░  close
  Rhythm                       0.58x  ███░░░░░░░░░░░░░░░░░  close
  Vocabulary                   0.56x  ███░░░░░░░░░░░░░░░░░  close
  Sentence shape               0.45x  ██░░░░░░░░░░░░░░░░░░  close
  Formatting                   0.27x  █░░░░░░░░░░░░░░░░░░░  close

Biggest differences   (each ▲ or ▼ is one standard deviation, up to 3)
                                                   this text   reference
  "but"                                              5.5 /1k     1.8 /1k   ▲▲
  "has"                                              7.4 /1k     2.5 /1k   ▲▲
  Sentences starting However / Moreover                 2.9%        0.0%   ▲▲  reference never varies
  Em dashes                                          1.8 /1k     0.0 /1k   ▲  reference never varies
  LLM marker words (delve, crucial)                  1.8 /1k     0.0 /1k   ▲  reference never varies
  "these"                                            1.8 /1k     0.0 /1k   ▲  reference never varies
  "by"                                               9.2 /1k     4.4 /1k   ▲
  "are"                                              0.0 /1k     4.3 /1k   ▼

Pass --all to see every metric.
