Skip to main content
Glama
milcho0604

velog-mcp

by milcho0604
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

벨로그 초안 수정

velog_update_draft
Destructive

Replace an existing draft entirely by providing id, title, and body. Omitted fields reset, so pass current values to preserve tags, slug, series, and thumbnail. Rejects published post IDs to avoid accidental unpublish.

Instructions

기존 초안을 통째로 교체한다. 부분 수정이 아니다. 생략한 필드는 유지되지 않고 초기화된다 — tags 를 안 주면 기존 태그가 전부 지워지고, url_slug 를 안 주면 제목에서 새로 만들어 주소가 바뀌며, series_id 를 안 주면 기존 시리즈 연결이 끊긴다. 그래서 수정 전에 velog_get_post 로 현재 값을 읽어 바꾸지 않을 필드도 그대로 다시 넘기는 것을 권한다. 발행된 글의 id 는 거부한다(비공개로 내려가는 사고 방지).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
idYes초안의 id (velog_list_drafts 로 확인)
bodyYes본문 전체 (마크다운). 부분 수정이 아니라 교체다. 문체: 사람이 쓴 글처럼 자연스럽게. 긴 줄표(—)와 가운뎃점(·)을 쓰지 말고 쉼표나 마침표로 끊을 것. "**하나.** ... **둘.**" 식 볼드 번호 나열 금지(산문이나 평범한 불릿으로). 문단은 2~3문장으로 짧게, 200자가 넘으면 쪼갠다. "이 글의 한계" 같은 부록 절을 만들지 말고 해당 문단 자리에 한 문장으로 녹일 것.
tagsNo
titleYes글 제목. 긴 줄표(—) 금지. 부제는 콜론이나 괄호, 짧은 하이픈으로.
url_slugNo
series_idNo
thumbnailNo썸네일 이미지 URL (http/https). 생략하면 본문 첫 이미지로 자동 설정한다. 자동 설정을 원하지 않으면 null 을 준다
series_nameNo시리즈 **이름**(id 대신). 저장 전에 내 시리즈에서 찾는다. 못 찾으면 저장하지 않고 목록을 알려준다

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed5 schema fields changedv0.9.2
    • changedInput schema / properties / body / description
      Previous value: -"본문 전체 (마크다운). 부분 수정이 아니라 교체다"New value: +"본문 전체 (마크다운). 부분 수정이 아니라 교체다. 문체: 사람이 쓴 글처럼 자연스럽게. 긴 줄표(—)와 가운뎃점(·)을 쓰지 말고 쉼표나 마침표로 끊을 것. \"**하나.** ... **둘.**\" 식 볼드 번호 나열 금지(산문이나 평범한 불릿으로). 문단은 2~3문장으로 짧게, 200자가 넘으면 쪼갠다. \"이 글의 한계\" 같은 부록 절을 만들지 말고 해당 문단 자리에 한 문장으로 녹일 것."
    • removedInput schema / properties / thumbnail / anyOf
      Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]
    • addedInput schema / properties / thumbnail / type
      Added value: +[
+  "string",
+  "null"
+]
    • addedInput schema / properties / title / description
      Added value: +"글 제목. 긴 줄표(—) 금지. 부제는 콜론이나 괄호, 짧은 하이픈으로."
    • addedInput schema / properties / title / maxLength
      Added value: +255
  2. Changed4 schema fields changedv0.6.0
    • addedInput schema / properties / series_name
      Added value: +{
+  "description": "시리즈 **이름**(id 대신). 저장 전에 내 시리즈에서 찾는다. 못 찾으면 저장하지 않고 목록을 알려준다",
+  "minLength": 1,
+  "type": "string"
+}
    • addedInput schema / properties / thumbnail / anyOf
      Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "type": "null"
+  }
+]
    • addedInput schema / properties / thumbnail / description
      Added value: +"썸네일 이미지 URL (http/https). 생략하면 본문 첫 이미지로 자동 설정한다. 자동 설정을 원하지 않으면 null 을 준다"
    • removedInput schema / properties / thumbnail / type
      Removed value: -"string"
  3. First observedv0.3.0

TDQS

A4.6/5.0
Behavior5/5

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

Beyond the destructiveHint annotation, the description details exactly what destruction occurs: omitted fields are reset, missing tags are deleted, missing url_slug is regenerated from the title changing the URL, and missing series_id breaks the series link. It also discloses the safety guard that published post IDs are rejected. This is rich behavioral context that the annotations alone would not provide.

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

Conciseness5/5

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

The description front-loads the most critical fact (wholesale replacement), then lists concrete consequences, then gives a preventive workflow, then mentions a safety constraint. Every sentence carries load-bearing information; nothing is filler. Despite covering several behavioral nuances, it remains compact and scannable.

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 destructive 8-parameter tool with no output schema, the description thoroughly covers behavioral risks and pre-conditions. It explains omitted-field semantics across multiple parameters and the published-post guard. The only notable gap is the absence of any mention of the return value or success/failure response, which is mildly important when no output schema exists.

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

Parameters4/5

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

With only 63% schema coverage, the description compensates by explaining the behavioral consequences of omitting key optional parameters: tags, url_slug, and series_id. It adds meaning beyond the schema by stating that omission resets these fields. It does not discuss thumbnail or series_name, but the schema already covers those reasonably, so the total compensation is solid though not exhaustive.

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

Purpose5/5

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

Description opens with a specific verb and resource: '기존 초안을 통째로 교체한다' (replace an existing draft wholesale), making the core action unambiguous. It further distinguishes itself from sibling update tools by rejecting published post IDs, which separates it from velog_update_post.

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?

The description gives clear when-not guidance: published post IDs are rejected, meaning this tool is only for drafts. It also recommends reading current values with velog_get_post before updating and passing unchanged fields, which is actionable usage context. However, it does not explicitly name velog_update_post as the alternative for published posts, so the routing is slightly less direct than ideal.

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

Deploy Server

Other Tools