Skip to main content
Glama

한국어 프롬프트 모호성 분석 및 개선 (V3 실험)

analyze_prompt

Analyzes short, ambiguous Korean requests using conversation and document context, resolves what can be clarified, asks only necessary questions, and returns a task-ready prompt.

Instructions

사용자의 짧고 모호한 한국어 요청을 이전 대화와 첨부 문서 문맥으로 분석해 모호성을 해결하고, 필요한 경우에만 명확화 질문을 반환하며, 그렇지 않으면 실제 작업에 사용할 수 있는 개선된 Prompt를 생성한다.

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": {
      "current_user_request": "<current_user_request>"
    },
    "name": "analyze_prompt"
  }
}

Input Schema

NameRequiredDescriptionDefault
answersNo
task_hintNo
document_contextNo첨부 문서 내용 또는 요약
previous_analysisNo
recent_conversationNo최근 대화 내역 (오래된 순서)
current_user_requestYes사용자의 현재 요청 원문

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

B3.4/5.0
Behavior3/5

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

With no annotations, the description carries the full burden, and it does disclose the important branch: clarification questions are returned only when needed, otherwise an improved prompt is generated. It does not state that the tool is side-effect-free, does not explain how the answers parameter re-enters the flow, and gives no cost, latency, or language-restriction caveats.

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?

Everything sits in one dense but coherent sentence with no filler, and the conditional outcome is stated up front rather than buried. The nested Korean clauses make it slightly harder to parse than a two-sentence split would, but nothing is wasted.

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

Completeness3/5

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

For a six-parameter tool with a nested object and no output schema, the description should explain the answers feedback loop and the shape of both return modes. It covers the core flow and context inputs adequately but leaves those gaps, making it minimally sufficient rather than complete.

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 50%, and the description loosely references several parameters: the current request, recent conversation, and attached document context. It adds no meaning for task_hint (the planning/writing enum), answers, or previous_analysis, so half the parameters remain unexplained in both schema and prose. Baseline 3 fits given partial coverage.

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?

The description names a specific verb and resource: it analyzes short, ambiguous Korean requests against conversation and document context, then either returns clarification questions or produces an improved prompt. That dual-output behavior makes the purpose unambiguous. It falls short of a 5 only because there are no siblings to differentiate against and no explicit statement of what the tool does not cover.

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

Usage Guidelines3/5

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

Usage is implied by 'short and ambiguous Korean requests,' which signals the input shape the tool is meant for. However, there is no explicit when-not guidance, no statement about well-formed requests, and no alternatives to consider. The 'only when necessary' branch describes internal logic rather than selecting between tools.

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