Bareun (바른) — Korean NLP & Spell Checking

맞춤법·띄어쓰기 교정

correct_grammar

Read-only

한국어 문장의 맞춤법·띄어쓰기를 교정합니다. 교정된 문장과 교정 블록(원문/교정문) 목록을 돌려줍니다. 원문을 다시 쓰지 않고 어긋난 자리만 고치므로 글의 문체·말투가 그대로 남습니다. 고친 자리마다 근거가 되는 한글맞춤법 조항(예: 제1장 총칙 제2항)과 설명·예시가 함께 실려 오므로, 교열 근거를 대야 하는 일에 적합합니다. 띄어쓰기가 전혀 없는 음성 인식 결과·자동 자막 원고도 복원하고 문장 경계를 찾아 문장 부호를 넣습니다.

Input Schema

Name	Required	Description
`text`	Yes	맞춤법/띄어쓰기를 교정할 한국어 텍스트(필수)
`encoding`	No	오프셋 계산 인코딩. utf8/utf16/utf32 중 하나. 기본값 utf32
`treat_as_title`	No	입력 문장을 기사 등의 제목으로 취급. 기본 false
`custom_dict_names`	No	적용할 사용자 사전 이름 목록(선택)
`disable_confusion`	No	혼동모델 비활성화. 기본 false
`disable_bias_check`	No	혐오·차별 표현(낮잡는 말·차별 어휘) 확인 블록 비활성화. 기본 false(= 켬)
`disable_vx_spacing`	No	보조 용언 띄어쓰기 비활성화. 기본 false
`enable_contested_fix`	No	논란 여지가 있는 고침 활성화(되세요·에 있어서·복수 -들 남용 등 지침 기반 고침). 기본 false
`disable_caret_spacing`	No	복합명사 분리 사전 적용 비활성화. 기본 false
`enable_sentence_check`	No	문장 단위 점검 활성화(호응·접속·성분 생략·번역투 피동). 기본 false
`disable_split_sentence`	No	문장 단위 분할 비활성화. 기본 false
`disable_typo_correction`	No	오타(오탈자) 교정 비활성화. 기본 false
`enable_cleanup_whitespace`	No	불필요한 공백 정리 활성화. 기본 false
`enable_public_term_refine`	No	공공 용어 순화 제안 활성화(어려운 한자어·외래어·일본어 투를 쉬운 말로, 확인 블록만). 기본 false
`enable_limited_punctuation`	No	제약된 구두점 사용 기능 활성화. 기본 false

Schema Changelog

Changes observed during successful MCP inspections.

Changed4 schema fields changedOct 31, 2026

addedInput schema / properties / disable_bias_check

Added value: +{
+  "description": "혐오·차별 표현(낮잡는 말·차별 어휘) 확인 블록 비활성화. 기본 false(= 켬)",
+  "type": "boolean"
+}

addedInput schema / properties / enable_contested_fix

Added value: +{
+  "description": "논란 여지가 있는 고침 활성화(되세요·에 있어서·복수 -들 남용 등 지침 기반 고침). 기본 false",
+  "type": "boolean"
+}

addedInput schema / properties / enable_public_term_refine

Added value: +{
+  "description": "공공 용어 순화 제안 활성화(어려운 한자어·외래어·일본어 투를 쉬운 말로, 확인 블록만). 기본 false",
+  "type": "boolean"
+}

changedInput schema / properties / enable_sentence_check / description

Previous value: -"문장 단위 점검 활성화. 기본 false"New value: +"문장 단위 점검 활성화(호응·접속·성분 생략·번역투 피동). 기본 false"

First observedOct 25, 2026

TDQS

A3.9/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint/openWorldHint, yet the description adds real behavioral detail: style and tone are preserved, output includes corrected sentence plus original/corrected blocks, and each fix carries a Hangul spelling rule clause with explanation and example. It omits the hate/discriminative-expression check block, which is a notable side output an agent would want to know about.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Four sentences, front-loaded with the core action and return shape, then distinguishing traits and the special ASR case. No filler, though the sentence about rule citations and the ASR sentence could be slightly tighter.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 15-parameter, no-output-schema tool, the description supplies the missing return-value picture (corrected text, correction blocks, rule evidence) and notes style preservation. The numerous toggle flags are left entirely to the schema, which is acceptable given 100% coverage.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% across all 15 parameters, so the schema carries the semantics. The description only loosely gestures at a few behaviors (spacing restoration, sentence boundary punctuation) without mapping them to the many enable_/disable_ flags, so baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('한국어 문장의 맞춤법·띄어쓰기를 교정합니다') and adds the distinctive trait that it only fixes erroneous spots rather than rewriting. It never names the sibling analysis tools, but its corrective role is naturally distinct from tokenize/analyze_syntax.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear usage context: suited to work where proofreading grounds must be presented, and explicitly covers the ASR/auto-caption case with no spacing at all. It does not state when to avoid the tool or point to a sibling alternative.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Bareun (바른) — Korean NLP & Spell Checking tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?