korean-spell-check — Claude Code skill: Korean proofreading
korean-spell-check — a Claude Code skill that checks Korean spelling and returns edits instead of a rewritten draft. Long input gets split into chunks before it goes out.
MIT ★ 880 updated 2026-09-03
npx skills add sangrokjung/claude-forge --skill korean-spell-check -a claude-codeWhat korean-spell-check does
The skill posts your Korean text to nara-speller.co.kr, a third-party service. There is no local dictionary behind it, so anything you would not paste into a public web form — client names, personal records, deal figures, credentials — does not belong in the request. The service describes itself as free for individuals and students on a non-commercial basis. Read its terms before pointing it at paid work, because the licence decides whether that use is allowed at all.
First step is mandatory: run npx -y @nomadamas/[email protected] instruct korean-spell-check and follow what it prints. The CLI detects the runtime and shows the matching instruction set. To see the bundled helper files, run npx -y @nomadamas/[email protected] files korean-spell-check. Node.js 18+ is required. No npx? Use the offline copy at .claude/skills/korean-spell-check/instruction.md instead.
Suggestions stay narrow: the skill points at what to change rather than rewriting the paragraph. A set of hard rules ships with it. No payment, message delivery, final submission, cancellation or public posting without the user's explicit approval right before the action. Credentials never get asked for, printed or stored in chat, files or shell arguments. CAPTCHA, identity-proofing, legal and e-signature boundaries stay untouched. For a repository holding sensitive text, that outbound request is a real cost, not a theoretical one.
What it helps with
- proofread Korean text and get a list of corrections
- chunk a long document so the spelling service accepts it
- catch typos in Korean UI strings before a release
- review a Korean draft or translation without a manual pass
- keep the safety rules intact when handing text to a third party
Example prompts for the agent
- Proofread
messages.ko.jsonand show me only the corrections - Check the Korean in
README.ko.md, split the long file into chunks - Find typos in the Korean strings in
src/i18n/ko.ts
How to install korean-spell-check
1Skills CLI (Vercel)
npx skills add sangrokjung/claude-forge --skill korean-spell-check -a claude-code2Manually: put the files in place
~/.claude/skills/korean-spell-check/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/korean-spell-check/
1Skills CLI (Vercel)
npx skills add sangrokjung/claude-forge --skill korean-spell-check -a codex2Manually: put the files in place
.agents/skills/korean-spell-check/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add sangrokjung/claude-forge --skill korean-spell-check -a cursor2Manually: put the files in place
.cursor/skills/korean-spell-check/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install sangrokjung/claude-forge skills/korean-spell-check2Manually: put the files in place
.github/skills/korean-spell-check/SKILL.md1Skills CLI (Vercel)
npx skills add sangrokjung/claude-forge --skill korean-spell-check -a gemini-cli2Platform CLI
gemini skills install https://github.com/sangrokjung/claude-forge --path skills/korean-spell-checkManually: put the files in place
.devin/skills/korean-spell-check/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about korean-spell-check
How do I install korean-spell-check in Claude Code?
Run npx skills add sangrokjung/claude-forge --skill korean-spell-check -a claude-code. The skill lands at ~/.claude/skills/korean-spell-check/SKILL.md.
What does korean-spell-check do?
It sends Korean text to nara-speller.co.kr, splits long input into chunks, and returns correction suggestions. The output focuses on the changes rather than a full rewrite.
Where does my text go during a check?
To the third-party service nara-speller.co.kr over a POST request. No local dictionary is involved, so client names, personal data and credentials should stay out of the request.
Source text
SKILL.md · MIT
--- name: korean-spell-check description: Use Nara/PNU Korean spell-check surfaces conservatively to proofread Korean text, chunk long input, and return change-focused correction suggestions. license: MIT metadata: category: writing locale: ko-KR phase: v1 --- # korean-spell-check <!-- k-skill:cli-stub — generated by scripts/generate-skill-stubs.js; edit skill.json / instruction.md instead --> > **Your text leaves this machine.** This skill checks spelling by POSTing the text to > `nara-speller.co.kr`, a third-party service; it is not a local dictionary. Do not send client > names, personal data, deal figures, credentials or anything else you would not paste into a > public website (`rules/korean-writing-quality.md` §5). > > The upstream service states it is free for **individuals and students, non-commercial use > only**. Check its terms before using it in commercial work. ## Get the full instructions (required first step) Run this and follow its output as the primary instructions for this skill: ```bash npx -y @nomadamas/[email protected] instruct korean-spell-check ``` The CLI detects the current runtime (Dolshoi vault/CloakBrowser vs generic) and prints only the applicable instructions, always up to date. Helper files bundled with the CLI are listed by: ```bash npx -y @nomadamas/[email protected] files korean-spell-check ``` If `npx` is unavailable, install Node.js 18+ or follow https://github.com/NomaDamas/k-skill#readme, or read the vendored offline copy at `.claude/skills/korean-spell-check/instruction.md` (audited 2026-08-11, commit 86096da). ## Hard rules even without the CLI - Never execute payment, message/email delivery, final submission, cancellation, or public posting without the user's explicit approval immediately beforehand. - Never ask for, print, or store plaintext credentials in chat, files, or shell arguments. - Never bypass legal, physical-presence, CAPTCHA, identity-proofing, or electronic-signature boundaries.
© sangrokjung · MIT · sangrokjung/claude-forge