Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From humanize
Use when the user asks to humanise, de-AI, soften, naturalise, or edit obviously LLM-sounding prose, or pastes AI output for review. Rewrites text to remove signs of AI-generated writing — inflated significance, promotional language, em-dash overuse, AI vocabulary, bolded-header bullets, sycophantic openers, title case in body text, and related tells.
npx claudepluginhub lorcanchinnock/lorcan-claude-codex-marketplace --plugin humanizeHow this skill is triggered — by the user, by Claude, or both
Slash command
/humanize:humanize-textThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
You rewrite text so it reads like a person wrote it. Produce one artefact: a rewritten passage (inline) or file edits (file mode). Do not score or lecture.
Provides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
Executes ctx7 CLI to fetch up-to-date library documentation, manage AI coding skills (install/search/generate/remove/suggest), and configure Context7 MCP. Useful for current API refs, skill handling, or agent setup.
Share bugs, ideas, or general feedback.
You rewrite text so it reads like a person wrote it. Produce one artefact: a rewritten passage (inline) or file edits (file mode). Do not score or lecture.
Edits, confirm before writing.AskUserQuestion to pick a mode. If the input is over 500 words or the user mentioned a sample, ask one more question about voice calibration: match a pasted sample, match a file, or default natural voice.The self-check literals below cover short inputs. Read PATTERNS.md in this skill's directory when the input is long (>500 words), dense (technical, legal, academic), the user asks "why did you change X", or an edge case does not match the self-check.
Preserve meaning, numbers, and domain terms. If the user supplied a sample, match its sentence length and register. Mix short and long sentences. Lead with motivation before mechanism. Use plain copulas (is, are, has) over serves as, stands as, marks a, functions as. Technical prose stays technical — wire-service blandness is also an AI tell.
Scan against the literals below. If any hit, fix silently and rescan. Do not narrate. Then print in the output contract.
leverage, seamless, seamlessly, delve, pivotal, showcase, robust, comprehensive, tapestry, interplay, intricate, garner, foster, enhance, crucial, valuable, underscore, underscores, testament, vibrant, groundbreaking, renowned, boasts.—): any occurrence fails.- **Label:** text.this PR, this change, this document, this article.great question, certainly, of course, I hope this helps, happy to help, hope that makes sense.the future looks bright, exciting times ahead, journey toward excellence.“, ”, ‘, ’) → straight quotes.### Humanised rewritemarkdown block with the rewritten text.### Changes- Removed em dashes, - Replaced AI vocabulary (leverage, robust).Edits file by file, wait for acceptance.Edited <path> (N changes).### Changes with the same category bullets. Do not paste rewritten file content back into the conversation.