AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

li-human — De-slop drafts & score AI tells

Strips em dashes, invisible Unicode and AI slop words from a draft, then scores it on a five-check human-sounding panel.

Content & WritingBeginner★ 2,878⑂ 456AI score 9/10Last updated: Sep 17, 2026

What it does

  • Automatic cleanup via humanize.py: deletes 17 classes of invisible characters (zero-width spaces/joiners, word joiners, soft hyphens, BOMs, Unicode tag chars), converts em dashes to commas, en dashes to hyphens, curly quotes to straight, ellipsis to three dots, and swaps 100+ stock AI phrases from slop.json (delve, leverage, robust, seamless, "in today's fast-paced world") for plain words — capitalisation preserved, URLs untouched.
  • Flags structural tells instead of rewriting them: "it's not just X, it's Y", rule-of-three triads, one-word rhetorical questions, rocket/fire/bulb emoji, hashtag walls, reflex engagement bait, uniform sentence and bullet lengths. You rewrite those lines by hand, then re-run the detector.
  • Scores five signals with detect.py: BURSTINESS, SPECIFICITY, SLOP DENSITY, FINGERPRINT, VOICE (0–100, higher is more human). Verdict weights the mean 60% and the weakest check 40%; PASS needs 70+ overall with nothing below 55.
  • Before/after proof: detect.py before.txt after.txt shows the delta.

Who it's for

  • Anyone posting on LinkedIn who wants drafts to stop sounding machine-generated
  • Content writers, marketers and ghostwriters editing AI-assisted copy
  • Students and freelancers sanity-checking text before submission (note: these are local heuristics, not GPTZero/Originality/Turnitin verdicts)

Example uses

  1. "Humanize this post" → python3 humanize.py draft.txt -o clean.txt --report and read the change log.
  2. "Does this sound like AI?" → python3 detect.py draft.txt to see the five scores and the weakest signal.
  3. "Remove the em dashes" → collapses — to , and cleans the leftover double punctuation.
  4. After hand-fixing flagged lines, python3 detect.py before.txt after.txt to show the improvement.

· · · 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/Jakeschincariol/linkedin-agent-skill/HEAD/skills/li-human/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)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. 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 AI Skill Mart.
Copy the skills/li-human folder from the GitHub repo Jakeschincariol/linkedin-agent-skill into my ~/.claude/skills/li-human/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/Jakeschincariol/linkedin-agent-skill.git /tmp/linkedin-agent-skill && mkdir -p ~/.claude/skills && cp -r /tmp/linkedin-agent-skill/skills/li-human ~/.claude/skills/

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

  1. Open a terminal (Terminal on macOS/Linux, WSL or Git Bash on Windows).
  2. Confirm Python 3 is available: python3 --version. Install from python.org if missing.
  3. Clone the repo: git clone https://github.com/Jakeschincariol/linkedin-agent-skill.git
  4. Create the skills folder: mkdir -p ~/.claude/skills
  5. Copy the skill in: cp -r linkedin-agent-skill/skills/li-human ~/.claude/skills/
  6. Verify with ls ~/.claude/skills/li-human — you should see SKILL.md, humanize.py, detect.py and slop.json.
  7. Restart Claude Code and try: "humanize this draft and score it".
  8. If the lexicon strips a word you genuinely use, open slop.json and delete that entry.
View source on GitHub ↗License: MIT