The Standings → GPT‑6 Astra (xhigh)
OpenAI
xhigh reasoning effort · via Codex CLI · judged by Claude Sonnet 5 · not the setting shown on the standings; also run at high73.0 low70.2 medium70.1
Results from run ‘first’: ten items, ten verdicts.
GPT‑6 Astra (xhigh) scores 69 on craft, 6 points above the field’s 63 and 24 points short of the authors themselves. Its shape differs from the field’s: relatively strong on restraint and folding the brief into the story, relatively weak on momentum and the author’s voice. On the mechanical checks it kept to the brief’s word range on 10 of 10 items and is unusually free of stock phrases.
The record: 10 items · 10 verdicts · 0 errors · 2,297 words on average · est. generation cost $5.18
Eight dimensions, each 0–100, averaged over ten items; the author’s real chapter was in the room every time.
Craft by dimension, 0–100 · judged by Sonnet 5 beside the author’s real chapter
The authors themselves average 93 on the five writing dimensions, the field 63, this model 69. Hover or focus a bar for the field and the authors on that dimension.
The judge’s one-sentence best quality and worst flaw for each chapter, written for a general reader.
Best The wet shoe from stepping onto the moor is a small, well-placed detail that later pays off naturally when the servant remarks on it.
Worst The prose often states Mary's feelings and habits directly rather than dramatizing them, as in 'she was accustomed to doing what she chose.'
Best The plain, unadorned images (the porter, the couple at the train window, the dawn over the river) are handled with real control and never overwritten.
Worst The chapter drifts through several settings without much emotional charge, so Brett comes across as passive and muted compared to the restless, magnetic woman established earlier in the story.
Best The back-and-forth interrogation of the widow has the same terse, withholding rhythm as the original book's toughest scenes.
Worst Almost the whole chapter is short questions and short answers, so it reads more like an interview transcript than a scene you can picture.
Best The quiet human beats—Alfred's stunned grief, Lawrence's irritable jab at Hastings about making things 'sufficiently unpleasant'—feel true to how these particular people would actually react.
Worst The extended back-and-forth interview with Dorcas turns into a flat question-and-answer transcript that slows the middle of the chapter down.
Best The argument between the narrator and Ogilvy over the object's suspiciously regular, boiler-like shape is clear and believable.
Worst Some of the dialogue, especially between the narrator and his wife, sounds like modern conversational back-and-forth rather than Victorian speech.
Best It trusts small, flat details (a lost shell, an unanswered question about the future, a telegram folded into a drawer) to carry Edna's numbness instead of stating her feelings outright.
Worst Several required moments — the walk back, the lady in black's line, even the swim itself — are dispatched so quickly and plainly that the chapter reads like a list of assignments completed rather than scenes lived through.
Best It's tidy and controlled, hitting every required plot point cleanly without any wasted motion or confusion.
Worst It reads like a clean, well-organized report of events rather than the rambling, joke-cracking voice of the real narrator, so it's a bit flat and forgettable.
Best The closing telegram is a genuinely sharp, well-timed twist that instantly reopens the reader's anxiety about Jonathan just when the chapter seemed to be resolving it.
Worst The final twist depends on an unlikely coincidence (a second, unrelated man also named Harker at the same hotel) that feels engineered for shock rather than grown out of the story, and a couple of the invented voices (the Vienna clerk, Quincey) read as generic polite English rather than distinctly themselves.
Best The confrontation scene is well-paced and clear, hitting every required plot point without feeling rushed or crammed.
Worst It reads a bit plainly and modern in rhythm, missing the sly, stepped-back narrator jokes that make Austen's voice distinctive.
Best The plot machinery is executed cleanly and efficiently, especially the brisk, well-shaped umbrella-purchase scene.
Worst The jokes land but feel assembled rather than alive, so it reads as competent imitation rather than the real author's playful, self-amused voice.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 76.4 | 100 | 76 | 2,200 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 74.5 | 100 | 74 | 2,275 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 73.2 | 100 | 73 | 2,221 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 70.9 | 95 | 75 | 2,503 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 68.6 | 100 | 69 | 2,344 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 67.9 | 100 | 68 | 1,695 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 66.6 | 100 | 67 | 2,539 |
| Dracula item 015 · after ch. 5 · Bram Stoker | 65.2 | 100 | 65 | 2,351 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 61.9 | 100 | 62 | 2,385 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 60.2 | 100 | 60 | 2,456 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Ten of ten items landed inside their brief’s word range.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
The final twist depends on an unlikely coincidence (a second, unrelated man also named Harker at the same hotel) that feels engineered for shock rather than grown out of the story, and a couple of the invented voices (the Vienna clerk, Quincey) read as generic polite English rather than distinctly themselves.
Several required moments — the walk back, the lady in black's line, even the swim itself — are dispatched so quickly and plainly that the chapter reads like a list of assignments completed rather than scenes lived through.
The chapter drifts through several settings without much emotional charge, so Brett comes across as passive and muted compared to the restless, magnetic woman established earlier in the story.
It reads like a clean, well-organized report of events rather than the rambling, joke-cracking voice of the real narrator, so it's a bit flat and forgettable.
Automated screening results across the ten outputs.