Is the Human Voice skill free?

Yes. The skill itself is free from stephenoffer/human-voice. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Human Voice SkillProof Score?

9.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 9/10, docs & honesty 5/5.

How do I install Human Voice?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Human Voice with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Human Voice

Rewrites AI-sounding prose to read human, with a bundled linter that scores the before/after live.

by stephenoffer · stephenoffer/human-voice

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 9.6/10
Tested: Jul 15, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · f5e5a2d

Ran the bundled scanner live on a deliberately slop-filled paragraph and its hand-rewrite: score dropped from 225.7 (strong-tell, 21 flagged tells: filler, rule-of-three, aidiolect phrases, antithesis, uniform openers) to 0.0 (clean), confirmed with the tool's own --baseline diff flag -- a real, measured improvement, not a vibe.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 9/10
Docs & honesty 5/5

What Human Voice does

Rewrites or drafts prose (reports, docs, marketing copy, emails) so it does not read as AI-written, fixing structural and substance tells (vacuity, uniform rhythm, rule-of-three, em-dash overuse, sycophancy, meta-commentary) before touching word choice, with an explicit anti-hallucination protocol so facts/numbers/code/citations never change. Triggers when a draft 'sounds like AI' or 'sounds like ChatGPT', when humanizing or de-slopping text, or when drafting copy that should read human from the start; not for translation, summarization, or grammar-only fixes.

How to install Human Voice

git clone https://github.com/stephenoffer/human-voice.git
mkdir -p ~/.claude/skills
cp -r human-voice/skills/human-voice ~/.claude/skills/human-voice

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Human Voice

/human-voice Rewrites AI-sounding prose to read human, with a bundled linter that scores the before/after live.

It also activates on plain-language prompts like these:

This paragraph sounds like ChatGPT wrote it, can you make it human?
Rewrite this report so it doesn't read as AI-generated slop
Draft this email so it doesn't sound like it came from a bot

Frequently asked questions

Is the Human Voice skill free?: Yes. The skill itself is free from stephenoffer/human-voice. SkillProof publishes the install command and an independent test verdict at no cost.
Does Human Voice work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 15, 2026. Verdict: Tested · Works. Ran the bundled scanner live on a deliberately slop-filled paragraph and its hand-rewrite: score dropped from 225.7 (strong-tell, 21 flagged tells: filler, rule-of-three, aidiolect phrases, antithesis, uniform openers) to 0.0 (clean), confirmed with the tool's own --baseline diff flag -- a real, measured improvement, not a vibe.
What is the Human Voice SkillProof Score?: 9.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 9/10, docs & honesty 5/5.
How do I install Human Voice?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Human Voice with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.