humanizer
Humanize AI-sounding text with natural rewrites, voice preservation, and editorial critique.
Skill by apoapostolov
Chalk Labs
Need help deploying humanizer?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/apoapostolov-humanizer
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 1, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/humanizer
What's inside
- Skill name
humanizer- Description
- Humanize AI-sounding text with natural rewrites, voice preservation, and editorial critique.
- License
- MIT
- Instructions
- 77 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- files: referenced but missing: references/examples.md
Files (5)
- SKILL.md
- SOURCE.md
- cover-letter.md
- humanizing-text.md · references/
- pattern-catalog.md · references/
SKILL.md (first lines)
# Humanizer Make text sound natural, specific, and true to its writer. Improve the writing; do not optimize for detector evasion or manufacture human-like mistakes. ## Establish the brief Identify the source text, task, audience, purpose, format, desired voice, and constraints. Protect facts, quotations, technical terms, supported uncertainty, and deliberate quirks. If no text is provided, ask for it. Ask one brief question only when missing context would materially change the result; otherwise make the smallest safe assumption and proceed. ## Choose the depth - **Cleanup:** Fix local stiffness, filler, repetition, and response residue. - **Rewrite:** Rebuild sentences and paragraphs while preserving meaning. - **Voice match:** Follow a supplied sample or clear voice brief. - **Critique:** Diagnose the strongest problems and recommend repairs without rewriting. Use the lightest intervention that solves the request. ## Load the right reference - Read [references/humanizing-text.md](references/humanizing-text.md) for full rewrites, voice or tone work, audience adaptation, fiction, dialogue, and long-form editing. It is the main structure for humanizing text. - Read [references/pattern-catalog.md](references/pattern-catalog.md) for a systematic audit, stubborn draft, or detailed diagnosis of recurring writing patterns. Use patterns as editorial prompts, never proof of authorship. - Read [references/examples.md](references/examples.md) when contrastive examples would improve calibration. Load only what the task needs. ## Edit 1. Preserve the draft's meaning, evidence, uncertainty, and recognizable voice. 2. Put the reader's needed fact, action, or decision before framing and ceremony. 3. Replace vague claims with supported specifics; never invent detail. 4. Let sentence rhythm, paragraph shape, and emphasis follow the thought. 5. Keep useful personality: stance, warmth, humor, restraint, tension, or roughness. 6. Remove template language, generic uplift, assistant chatter, placeholders, abrupt cutoffs, and mismatched formatting. 7. Compare the result with the source and its surrounding text before delivery. ## Guardrails - Do not treat polished grammar, a dash, a triad, passive voice, or any single feature as inherently artificial. - Preserve intentional rhetoric, genre conventions, accessibility, and house style when they serve the text. - Do not rewrite quoted examples, code blocks, or attributed excerpts unless the user asks. Flag a problem in protected material separately when it matters. - Do not add fake sources, facts, quotations, memories, emotions, sensory detail, slang, typos, or first-person experience. - Do not flatten necessary technical precision or evidence-based qualification. - Flag misleading or manipulative claims instead of polishing them into stronger deception. ## Deliver Follow the requested format. If none is given: - For a rewrite, return the revised text first. Add a note only for a material choice, ambiguity, or factual concern. - For a critique, name the strongest clusters, cite short examples, and prescribe specific repairs. - For a mixed request, give a compact diagnosis followed by the rewrite. Return only what helps the user. Do not add generic offers or commentary around a clean rewrite.
From apoapostolov/MuhFweeCeeVee-Professional-CV-Template-Generator, MIT.
More from apoapostolov/MuhFweeCeeVee-Professional-CV-Template-Generator
- 0.603score
muhfweeceevee-development
apoapostolov
Use when developing, debugging, or extending MuhFweeCeeVee (MyFreeCeeVee) — the local-first CV composer at…
SkillAFreshness: A★ 2Claude Code +9 - 0.603score
cv-review-lab
apoapostolov
Use when reviewing or implementing MuhFweeCeeVee workflows for ATS keyword coverage, natural resume language, AI-writing detector comparisons, PDF overflow…
SkillAFreshness: A★ 2Claude Code +9 - 0.593score
next-dev-workflow
apoapostolov
Use when Next.js dev server on WSL shows stale UI, HMR misses, or wrong port after editing MuhFweeCeeVee.
SkillAFreshness: A★ 2Claude Code +9
Signals
- Stars
- 2
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Jul 26, 2026 (3 months ago)
- Last release
- Sep 30, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from apoapostolov
- 0.603score
muhfweeceevee-development
apoapostolov
Use when developing, debugging, or extending MuhFweeCeeVee (MyFreeCeeVee) — the local-first CV composer at…
SkillAFreshness: A★ 2Claude Code +9 - 0.603score
cv-review-lab
apoapostolov
Use when reviewing or implementing MuhFweeCeeVee workflows for ATS keyword coverage, natural resume language, AI-writing detector comparisons, PDF overflow…
SkillAFreshness: A★ 2Claude Code +9 - 0.593score
next-dev-workflow
apoapostolov
Use when Next.js dev server on WSL shows stale UI, HMR misses, or wrong port after editing MuhFweeCeeVee.
SkillAFreshness: A★ 2Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get humanizer running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
humanizer skill. Agent Mart. https://agentmart.dev/skills/apoapostolov-humanizer (last updated 2026-10-24)