Remove AI writing patterns from prose. Use when drafting, editing, or reviewing text to eliminate predictable AI tells.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Humanize
Eliminate predictable AI writing patterns from prose.
## Scope
Rewrite phrasing, not meaning. Preserve facts, numbers, file paths, key names,
API names, links, claims, uncertainty, scope, modality, and intentional voice.
If a cleaner sentence changes what the document asserts, leave the sentence
unchanged.
Identify the document type before editing. Prose, marketing copy, technical
docs, policies, changelogs, licenses, and agent instructions need different
levels of compression and different registers.
Do not rewrite code blocks, commands, headings, tables, links, or quoted text
unless explicitly asked. Preserve intentional fragments, repetition, parallel
structure, and rhetorical phrasing when they improve navigation or emphasis.
## Core Rules
1. **Cut filler phrases.** Remove throat-clearing openers, emphasis crutches,
generic transitions, and empty intensifiers. Keep adverbs that carry a fact,
such as `locally`, `manually`, `asynchronously`, or `optionally`. See
[references/phrases.md](references/phrases.md).
2. **Break formulaic structures.** Avoid binary contrasts, negative listings,
…
The line at the top of this page pins the label git:20261004.03effb8, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
GET https://markdownregistry.com/api/v1/artifacts/art_fztwurdpypum75ih
GET https://markdownregistry.com/api/v1/resolve?ref=mithunwijayasiri/dotclaude/humanize
GET https://markdownregistry.com/api/v1/blob/16bb115abb171277239b1ef8c6dae6dfa35a4147c8ec4343f669f17b8617561c
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
mithunwijayasiri/dotclaude · skills/ask-first/SKILL.md · Ask the user clarifying questions before acting, instead of filling gaps with guesses. Use when starting to implement a…
mithunwijayasiri/dotclaude · skills/claude-project-rules/SKILL.md · Author Claude Code rule files (.claude/rules/*.md). Use when asked to create, add, edit, or review a project or user…
mithunwijayasiri/dotclaude · skills/compact-markdown/SKILL.md · Writing style for markdown files. Strips filler, writes telegraphic and content-only — docs are for AI agents, not…
mithunwijayasiri/dotclaude · skills/empty-commit/SKILL.md · Use when a CI pipeline failed on something unrelated to the code and the PR is blocked — pushes an empty commit to…
mithunwijayasiri/dotclaude · skills/git-merge/SKILL.md · Merge one branch (usually the default branch) into the current branch while preserving local unstaged changes…
mithunwijayasiri/dotclaude · skills/git-worktree/SKILL.md · Use when the user asks to create, enter/switch to, or remove a git worktree on Windows — makes it under…
mithunwijayasiri/dotclaude · skills/github-issue-creator/SKILL.md · Create or edit GitHub issues from raw notes, error logs, voice dictation or screenshots, following the repo's issue…
mithunwijayasiri/dotclaude · skills/handoff-session/SKILL.md · Use when the user wants to hand off, continue in a new session, or pass the current work to another agent. Compacts the…
mithunwijayasiri/dotclaude · skills/offload-task/SKILL.md · Write one stuck problem into a standalone markdown brief so a fresh session in any coding agent can attack it without…
epoko77-ai/im-not-ai · skills/humanize/SKILL.md · AI가 쓴 한글 텍스트를 자연스럽게 윤문하는 진입 명령. humanize-korean 파이프라인을 shim 의 route_hint 가 정한 경로(light 1콜 · standard 2콜 · heavy 3+콜)로…
pedrohcgs/claude-code-my-workflow · .claude/skills/humanize/SKILL.md · Read-only audit of `.tex`, `.qmd`, or `.md` text for AI-voice tells — boilerplate transitions ("Moreover"…
fcakyon/claude-codex-settings · plugins/humanize/skills/humanize/SKILL.md · This skill should be used when writing or revising text, including responses, documentation, comments, emails, posts…
aperivue/medsci-skills · skills/humanize/SKILL.md · Use when a manuscript or response-to-reviewers letter reads as AI-written. Scans for 27 AI writing patterns and…
mathews-tom/armory · skills/humanize/SKILL.md · Detects and removes AI-generated writing patterns while preserving meaning and facts, using a deterministic scan and…
uchicago-dsi/ai-sci-skills · skills/humanize/SKILL.md · Edit prose so it sounds like the author and reads plainly - stripping formulaic AI phrasing and replacing jargon with…
kalyvask/winning-writing · skills/humanize/SKILL.md · Makes a draft read like a real human typed it — adds asymmetry, light contractions, and (rarely) one safe typo. Use…
