humanize

/humanize — Remove the Korean AI feel

humanize-korean triggers the skill and runs polishing on the Korean text (or file) passed as an argument.

Input

$ARGUMENTS

Behavior

1. If the argument is empty: guide with "Please paste the text to polish" and exit.
2. If the argument is a file path (.txt/.md), Readto load the body.
3. If the argument is text, use it as input as is.
4. humanize-korean Follow the skill SKILL.md procedure (Phase 0 → deliver results) — the path is determined by route_hint(light|standard|heavy), and --strict if , heavy is fixed.
5. Deliver results:
   • One-line status (change rate / grade / self-verification passed)
   • Polished body text (markdown block)
   • Detection counts by category before/after
   • 3–5 highlights of major changes
   • If grade B or lower, guide with "/humanize-redocan be used for a second polishing pass"

Options (in natural language at the end of the argument)

• 장르: 칼럼|리포트|블로그|공적 — specify genre (if omitted, automatically inferred from the first 300 characters)
• 강도: 보수|기본|적극 — polishing intensity (default: default)
• 최소심각도: S1|S2|S3 — detection threshold (default: S2)
• --strict — force precise 3 calls (diagnosis → targeted polishing → finalize)

Notes

• Classification system: humanize-korean/references/ai-tell-taxonomy.md
• Polishing prescription: humanize-korean/references/rewriting-playbook.md