Quirón 1.2.0

The skill now builds a rewrite from the writer's notes. The post follows the notes and drops the tutorial scaffolding the draft added around them, like a "basics" section or a closing summary. Sections end on their last fact and the rewrite leaves no placeholder from the draft behind. The skill also names the sincerity markers Claude adds: earlier rewrites put honestly in a third of posts, and genuinely now counts in the AI lexicon.

After the loop converges, a fresh agent reads the text cold, once, and the rewriter fixes the reasons that point at a sentence. Stories get their own rules in references/fiction.md, about the story's shape: callbacks and bookends, a clockwork arc and a moral stated outright.

Measured with blind Claude Opus judges who read each text alone (human originals score 4 to 8): blog posts went from a mean P(AI) of 47.8 to 34.9 on 20 titles, better on every one. The first of these changes took held-out titles from 36.7 to 31.8. Spanish posts went from 38.8 to 29.2, held-out stories from 74.0 to 69.7, and rewrites by GPT-5.6 through Codex from 67.0 to 55.3. The harness behind these numbers is in eval/e2e/, and references/numbers.md records them along with the candidates that were measured and left out.

Quirón 1.1.0

Adds a review step with the writer. After the rewrite loop converges, the skill asks about negated setups one sentence at a time: a short "X isn't Y." that opens a paragraph or sits right before the point, like "What surprised me most wasn't a number." Each question gives the reasons a reader may take the sentence for AI and a suggested change. The writer can keep it, use the suggestion, delete it, or write their own. The same list is available from scripts/audit.py --review.

The rewrite loop never changes these sentences on its own. Fixing them automatically made rewrites read more AI to blind judges (mean P(AI) 43.6 to 46.4 over 20 titles). With the review step, rewrites measure the same as 1.0.0 (43.7). The checklist, T1 and check.sh are unchanged, and references/numbers.md now records three checks that were measured and left out.

Quirón 1.0.0

First release of Quirón, an AI humanizer skill for Claude Code, Codex and Cursor. It removes AI slop and the signs of AI writing from a draft, then measures the rewrite against human writing published before ChatGPT.

Install with npx skills add ilien-dev/quiron, or inside Claude Code with /plugin marketplace add ilien-dev/quiron and /plugin install quiron@quiron.