Clean User-Facing Text

A final hygiene pass that strips suspicious invisible Unicode and rewrites prose for clarity while preserving facts, formatting, and the writer's voice.

Content & WritingIntermediate★ 767⑂ 73AI score 10/10Last updated: Aug 23, 2026

What it does

Deterministic Unicode pass: uses the bundled inspect_text.py and clean_text.py to detect and remove zero-width characters, odd spaces, and directional controls, with an inspect → clean → re-inspect loop and stats.
Voice-preserving rewrite: varies clause order, sentence boundaries, rhythm, and connectors, and drops formulaic transitions — while keeping every claim, number, name, citation, and required disclosure intact.
Non-prose protection: fenced/inline code, commands, paths, URLs, identifiers, API names, formulas, and verbatim quotes stay untouched; no variable renaming or code reformatting.
Honest reporting: audits separate what is verifiable (characters removed), best-effort (prose rewriting), and not established (detector evasion, proof of human authorship).

Who it's for

Writers and editors finalizing articles, manuscripts, reports, docs, product copy, or UI strings.
Developers whose diffs or greps break because pasted text carries invisible characters.
Anyone editing Markdown/HTML who needs prose polished without touching embedded code.

Limits: binary containers (PDF, DOCX, images, archives) are unsupported, and the skill explicitly refuses authorship-evasion use cases.

Examples

"Tighten the prose in this README but leave every code block byte-for-byte" → prose-only rewrite, then the Unicode pass on the file.
"Check draft.md for hidden characters" → inspect_text.py --json reports hit positions; hits inside code spans are left alone instead of a whole-file clean.
"Give me an audit of the report" → counts of removed characters, rewriting flagged as best-effort, and no claims about beating AI detectors.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/ShadowAqueduct/watermark-remover/HEAD/skills/clean-user-facing-text/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)

Download the ZIP with the button below.
In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.

↓ Download ZIP

Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/clean-user-facing-text folder from the GitHub repo ShadowAqueduct/watermark-remover into my ~/.claude/skills/clean-user-facing-text/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/ShadowAqueduct/watermark-remover.git && mkdir -p ~/.claude/skills && cp -r watermark-remover/skills/clean-user-facing-text ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

Open a terminal (Terminal on macOS/Linux, PowerShell or Git Bash on Windows).
Verify Python 3 is available: python3 --version (or py --version on Windows).
Clone the repository: git clone https://github.com/ShadowAqueduct/watermark-remover.git
Create the skills directory: mkdir -p ~/.claude/skills
Copy the skill in: cp -r watermark-remover/skills/clean-user-facing-text ~/.claude/skills/
Confirm scripts/inspect_text.py and scripts/clean_text.py exist inside the copied folder.
Restart Claude Code and try a prompt like "Clean the hidden Unicode in this file and polish the prose, leaving code untouched."

View source on GitHub ↗License: MIT