Writing Prose Like A Human
Strips AI writing tells: significance inflation, inflated verbs, trailing -ing editorializing, rule-of-three.
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 16, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 41a1564
Baseline about-page paragraph, written without the skill, landed on almost every word in the skill's own watchlist ('cherished cornerstone', 'testament', 'nestled in the heart of', 'vibrant hub') -- the skill-guided rewrite following the five rules replaced it with dated, concrete, specific sentences and zero watchlist hits, a stark and directly falsifiable before/after.
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 Writing Prose Like A Human does
A vocabulary watchlist and five concrete editing rules (be specific not significant, plain verbs, end sentences at the fact, vary rhythm, earn every adjective) for making prose read as human-written rather than LLM-generated, adapted from Wikipedia's 'Signs of AI writing' guide. Triggers when writing or editing prose, documentation, READMEs, or PR descriptions where the output must not read like AI content; ships with a companion prose-humanizer subagent that applies the rules in-place.
How to install Writing Prose Like A Human
git clone https://github.com/kylehughes/writing-prose-like-a-human-for-agents.git
mkdir -p ~/.claude/skills
cp -r writing-prose-like-a-human-for-agents/skills/writing-prose-like-a-human ~/.claude/skills/writing-prose-like-a-human
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Writing Prose Like A Human
-
/writing-prose-like-a-humanStrips AI writing tells: significance inflation, inflated verbs, trailing -ing editorializing, rule-of-three.
It also activates on plain-language prompts like these:
-
Rewrite this bakery about-page paragraph so it sounds less AI-written -
Edit my README so it doesn't read like ChatGPT wrote it -
Check this PR description for inflated verbs and rule-of-three filler
Frequently asked questions
- Is the Writing Prose Like A Human skill free?
- Yes. The skill itself is free from kylehughes/writing-prose-like-a-human-for-agents. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Writing Prose Like A Human work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 16, 2026. Verdict: Tested · Works. Baseline about-page paragraph, written without the skill, landed on almost every word in the skill's own watchlist ('cherished cornerstone', 'testament', 'nestled in the heart of', 'vibrant hub') -- the skill-guided rewrite following the five rules replaced it with dated, concrete, specific sentences and zero watchlist hits, a stark and directly falsifiable before/after.
- What is the Writing Prose Like A Human 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 Writing Prose Like A Human?
- 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 Writing Prose Like A Human 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.