Skip to main content
Glama

인용 일괄 점검

check_citations
Read-onlyIdempotent

답변 초안을 통째로 넣으면 그 안의 인용을 모두 뽑아 회계위키 원문과 대조: 기준서 문단 인용('제1116호 문단 46A·46B', '감사기준서 240 문단 A31', '제13장 문단 13.35')과 질의회신·감리사례 번호(SSI-38683, 2020-I-KQA008, KICPA-2025-30)가 실제로 있는지, 큰따옴표·낫표 안 인용문이 원문과 같은지 확인한다. 없는 문단에는 비슷한 번호를, 다른 인용문에는 가장 가까운 원문을 준다. 회계 질문에 대한 답변을 이용자에게 내보내기 전에 한 번 부를 것. 인용한 문단·문서가 있는지와 문구 일치만 보며, 그 문단이 질문에 맞는 근거인지는 판단하지 않는다. 넣은 초안은 저장하지 않는다. 결과는 이용자 질의에 대한 조회 목적으로만 제공되며 원문 저작권은 제정·발행기관에 있다. 인용 시 응답의 출처 URL 을 표기할 것.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "text": "<text>"
    },
    "name": "check_citations"
  }
}

Input Schema

NameRequiredDescriptionDefault
textYes점검할 답변 초안 (앞 20,000자까지 봄). 문단 인용은 기준서 표시와 '문단'을 함께 적은 것만 인식한다
domainNo기준서 표시 없이 문단 번호만 적힌 인용을 해석할 때의 기본 도메인 (생략 가능)

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / domain / enum
      Previous value: -[
      -  "ifrs",
      -  "kgaap",
      -  "audit",
      -  "icfr"
      -]New value: +[
      +  "ifrs",
      +  "kgaap",
      +  "audit",
      +  "icfr",
      +  "law"
      +]
  2. Added

TDQS

A4.1/5.0
Behavior4/5

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

Goes beyond the readOnly/idempotent annotations by disclosing that missing paragraphs yield a similar-number suggestion and mismatched quotes yield the nearest original text, and that the submitted draft is not stored. It also sets the copyright/attribution obligation, which the annotations cannot convey. It does not describe result shape or limits on the suggestions, keeping it short of a 5.

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?

Dense but front-loaded: what it extracts and checks comes first, then call timing, then scope limits and non-persistence. The trailing copyright/attribution sentence is useful policy context rather than padding. Slightly long relative to the two-parameter surface, but no sentence is redundant.

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 two-parameter, annotation-covered, no-output-schema tool this covers the essentials: input expectation, what is verified, what is not verified, that the draft is discarded, and what corrective output to expect. The absence of an output schema means return details (per-citation status fields, ordering, limits) are only sketched, which is the remaining gap.

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 description coverage is 100%, so the text (20,000-char truncation, paragraph-detection rule) and domain enum are already documented in the schema. The description only implicitly references the draft input ('답변 초안을 통째로 넣으면') and says nothing extra about how domain affects interpretation beyond what the schema states. Baseline 3 applies.

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?

States a specific verb+resource (extract all citations from a draft and cross-check them against 회계위키 originals) and enumerates exactly which citation forms are covered (기준서 문단, 질의회신·감리사례 번호) plus the two checks performed (existence, verbatim quote match). An agent knows precisely what this returns versus a generic search or fetch tool.

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 explicit call timing ('회계 질문에 대한 답변을 이용자에게 내보내기 전에 한 번 부를 것') and an explicit exclusion ('그 문단이 질문에 맞는 근거인지는 판단하지 않는다'), which tells the agent what this tool will not do. However, it never names a sibling (e.g. verify_quote or find_evidence) or states the condition that would select one of those instead, so alternative routing is left to inference.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources