Skip to content
SkillScript-led983 tokens in use

li-human

Removes invisible characters, em dashes and stock AI vocabulary from a draft with two local Python scripts, then scores it on a five-check panel.

3,760 stars on the repository MIT LicenseJakeschincariol/linkedin-agent-skill · skills/li-human
/plugin marketplace add Jakeschincariol/linkedin-agent-skill /plugin install linkedin-agent
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing li-human costs

95tokens idle
983tokens in use
5,789tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Script-led

Script-ledMixedModel-led

Most of this skill is scripts, so it behaves much the same whichever model runs it.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: detect.py, humanize.py.

Touches

shelldetectedwrites filesdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter12/15
  • Structure17/25
  • Examples6/15
  • Output10/15
  • Size15/15
  • Resources15/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 18
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    humanize.py, line 247

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Needs attention
  • Development activity28
  • Maintenance4
  • Community52
  • Maturity16
  • Momentum40

li-human cleans a draft with humanize.py and scores it with detect.py, both Python scripts with no dependencies that run on your machine. The cleaner deletes invisible characters, converts em dashes, curly quotes and ellipses, and swaps stock words and phrases from an editable lexicon.

Structural tells such as the not-only-but-also pattern, rule-of-three lists, hashtag walls and engagement bait are flagged for you to rewrite. The scorer rates burstiness, specificity, slop density, fingerprint and voice and gives a verdict of pass, review or flagged.

What You Get

  • A cleaned draft with a report of every invisible character, typographic change and replaced word
  • A list of structural tells flagged for you to rewrite by hand
  • A human score from 0 to 100 across five checks with a pass, review or flagged verdict
  • A before and after comparison that shows the change in score between two drafts

Common Use Cases

  • Cleaning a draft before it goes on LinkedIn
  • Checking whether text carries zero-width characters from copy and paste
  • Removing em dashes and stock phrases from any writing
  • Comparing a draft with its cleaned version to see what moved the score

What Makes It Unique

It runs real scripts, so cleaning and scoring are repeatable and measured by code. The word list lives in an editable slop.json file, and the scorer weights the weakest single check at 40 percent because a detector needs only one signal. Its instructions also limit the claim: the checks are local heuristics, they do not call any detector service and the skill never tells you text is undetectable.

Under The Hood

The skill tells the agent to run python3 humanize.py and python3 detect.py from its folder on your draft files, so it runs shell commands and can write a cleaned output file you name. Both scripts read slop.json, the lexicon, and work entirely on your machine with no network calls and no uploads. They use only the Python standard library. The agent then rewrites the flagged lines itself and re-runs the scorer.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers