Skip to content
SkillModel-led1,200 tokens in use

commit-ko

Smooths AI-written Korean commit messages into the short wording Korean developers use, keeping the type, scope, issue numbers and meaning.

6,692 stars on the repository MIT Licenseepoko77-ai/im-not-ai · extras/skills/commit-ko
./install.sh --extras
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing commit-ko costs

277tokens idle
1,200tokens in use
3,271tokens, 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 Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

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

Touches

shelldetected

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 →
Fair
  • Frontmatter8/15
  • Structure16/25
  • Examples3/15
  • Output0/15
  • Size9/15
  • 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

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community64
  • Maturity28
  • Momentum40

commit-ko takes a Korean commit message drafted by an AI tool and removes stiff office wording such as verbs that mean carry out or proceed with, noun-heavy phrases, over-polite endings and sentence shapes copied from English. The result is the short form Korean developers write.

It works on one or two lines at a time with a single model call. It reads one lexicon file of replacements, changes only the patterns it finds, and says so when a message already reads well.

What You Get

  • A cleaned commit message in the same shape as the draft
  • A one-line reason for each change
  • One or two alternative wordings when a case is unclear
  • A notice that the message is already natural when nothing needs changing

Common Use Cases

  • Cleaning up commit messages that an AI assistant drafted in Korean
  • Shortening messages that end in formal verbs before pushing
  • Checking a batch of commit messages for translated-sounding phrasing

What Makes It Unique

It is made for the commit message register, where a noun-form ending is the norm and polite endings sound off. It leaves loan words such as update, refactoring and bug alone because developers use them, and it treats English-style sentence structure, not word origin, as the problem. It does not invent reasons or effects that the draft never mentioned.

Under The Hood

The skill is an instruction file plus a lexicon reference. The agent reads the lexicon and rewrites the message in a single step. A small checker script ships with the repository to compare the lexicon header with the main pattern catalogue, but the skill itself does not run it. It does not fetch web pages, call APIs or write files.

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