Text Humanizer
Rewrites AI-sounding drafts into human prose without losing a single fact.
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 11, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Jul 16, 2026
⚠ This skill is no longer available upstream. Our re-check on Aug 10, 2026 couldn't find it any more (SKILL.md not found in repo tree). The test below is what we measured on Jul 11, 2026 and we're leaving it up as a record — but there is nothing left to install, so we've removed the command.
Our own skill — tested under the same protocol, disclosed. The bench ran 4 SKILL.md revisions against an unchanged naive-prompt baseline over 12 AI-generated texts; v1–v3 all lost the blind human-ness preference (8-0, 12-0, 12-0) before v4 inverted the method (freeze facts, rebuild structure freely) and won 8-4 with 129/130 claims intact and the fewest residual tells (2) of any arm. Caveats disclosed by the bench itself: single in-context LLM judge from the rewriter's own model family (no external detector reachable), one run, and a margin with several near-coin-flip texts — a clear but modest win, not a blowout. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.
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 7/10
- Docs & honesty 5/5
What Text Humanizer does
Detects and removes the structural tells of AI-written text — topic-sentence openers, rule-of-three lists, stock closers — by rebuilding the passage in a human voice instead of swapping vocabulary. A hard claim-freeze rule keeps every number, name, quote, and caveat intact through the rewrite, verified by a mandatory final re-read against the source.
Text Humanizer before / after — same task, with and without
Without the skill
In the ever-evolving landscape of software development, Docker has emerged as a pivotal technology that has fundamentally transformed how applications are built, shipped, and deployed. But what exactly is a container, and why has it become so crucial to modern engineering teams? At its core, a Docker container is a lightweight, standalone, executable package that includes everything an application needs to run... In conclusion, Docker containers offer a seamless, efficient, and portable approach to application deployment.
With Text Humanizer
A Docker container is a lightweight, standalone package that bundles everything an application needs to run — code, runtime, system tools, libraries, and settings. Unlike a virtual machine, which virtualizes an entire operating system, a container shares the host machine's OS kernel... Between the efficiency and the consistency, it's easy to see why Docker changed how teams build, ship, and deploy software: less overhead, fewer environment surprises, a more reliable path from code to production.
How to install Text Humanizer
Nothing to install: the source repository no longer has this skill. If the author brings it back, our daily re-check will pick it up and the command will reappear here.
Commands — how to trigger Text Humanizer
-
/text-humanizerRewrites AI-sounding drafts into human prose without losing a single fact.
It also activates on plain-language prompts like these:
-
Can you humanize this blog draft? It reads like ChatGPT wrote it. -
This client email sounds too AI-generated, can you make it sound more natural? -
I'm worried this product description will get flagged by an AI detector — clean it up but keep every claim exactly the same.
Frequently asked questions
- Is the Text Humanizer skill free?
- Yes. The skill itself is free from Skillproofdev/text-humanizer. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Text Humanizer work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 11, 2026. Verdict: Tested · Works. Our own skill — tested under the same protocol, disclosed. The bench ran 4 SKILL.md revisions against an unchanged naive-prompt baseline over 12 AI-generated texts; v1–v3 all lost the blind human-ness preference (8-0, 12-0, 12-0) before v4 inverted the method (freeze facts, rebuild structure freely) and won 8-4 with 129/130 claims intact and the fewest residual tells (2) of any arm. Caveats disclosed by the bench itself: single in-context LLM judge from the rewriter's own model family (no external detector reachable), one run, and a margin with several near-coin-flip texts — a clear but modest win, not a blowout. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.
- What is the Text Humanizer SkillProof Score?
- 8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 5/5.
- How do I install Text Humanizer?
- 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 Text Humanizer 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.