mcp-narajangteo

Name: mcp-narajangteo
Author: Datajang

by Datajang

Overview Schema Source Related Servers Discussions Admin

Python

Nara MCP Server (나라장터 입찰공고 검색)

MCP server for searching Korean government procurement bid notices from G2B (나라장터 - Nara Jangteo).

Features

🔍 통합 검색: 용역 입찰공고 + 사전규격을 키워드로 검색 (검색 기간 자유 설정, 기본 7일)
💰 예산 정보: 모든 검색 결과에 예산 금액 표시
📅 마감 상태 표시: 모든 공고를 표시하고 ✅ 진행중 / 🔴 마감 상태를 명확히 구분
📎 파일 추출: 제안요청서(RFP) 자동 다운로드 및 텍스트 추출
🗂️ 스마트 필터링: 제안요청서/과업지시서 파일만 자동 선별
🏢 맞춤형 추천: 부서 프로필 기반 유연한 추천 (Top N 또는 전체 목록)
📄 다형식 지원: HWP, HWPX, PDF, DOCX, XLSX, ZIP 파일 자동 처리
🎯 전략 분석: 첨부파일 기반 입찰 전략 제안

Related MCP server: Enterprise Operation Insight MCP Server

Quick Start

Get started in 3 steps:

Install from PyPI
```
pip install nara-mcp-server
```
Get API keys (2개 필요)
- 공공데이터포털 에서 "조달청_나라장터 입찰공고정보서비스" 활용신청 → NARA_API_KEY
- 동일 포털에서 "조달청_나라장터 사전규격정보서비스(HrcspSsstndrdInfoService)" 활용신청 → NARA_PRESPEC_API_KEY

Configure Claude Desktop Add to claude_desktop_config.json:

{
  "mcpServers": {
    "nara-jangteo": {
      "command": "uvx",
    "args": [
        "--python",
        "3.11",
        "--from",
        "nara-mcp-server",
        "nara-server"
      ],
      "env": {
        "NARA_API_KEY": "입찰공고 API KEY",
        "NARA_PRESPEC_API_KEY": "사전규격 API KEY",
        "UV_LINK_MODE": "copy"
      }
    }
  }
}

Start using! Restart Claude Desktop and ask: "나라장터에서 'AI' 키워드로 입찰공고를 검색해줘"

Prerequisites

1. API 키 발급 (필수 — 2가지)

나라장터 API는 일반 입찰공고와 사전규격이 별도 서비스로 분리되어 있어 각각 활용신청이 필요합니다.

① 입찰공고 API 키 (NARA_API_KEY)

공공데이터포털 접속 및 회원가입
"조달청_나라장터 입찰공고정보서비스" 검색 후 활용신청
마이페이지 > 개발계정 에서 ServiceKey 확인 (일반 인증키 Decoding 사용)

② 사전규격 API 키 (NARA_PRESPEC_API_KEY)

동일 포털에서 "조달청_나라장터 사전규격정보서비스(HrcspSsstndrdInfoService)" 검색 후 활용신청
승인 후 동일하게 ServiceKey 확인

두 서비스의 키가 동일할 수도 있으나, 각각 활용신청이 완료되어야 정상 동작합니다.

2. Python 환경

Python 3.10 이상 필요

Installation

Option 1: From PyPI (Recommended)

The simplest way to install:

pip install nara-mcp-server

Note: This installs the nara-server command globally for easy access.

Option 2: From Source (For Development)

If you want to contribute or modify the code:

git clone https://github.com/Datajang/narajangteo_mcp_server.git
cd narajangteo_mcp_server
pip install -e .

Configuration

Using .env File (Recommended for Development)

Create a .env file in the project root:

# .env
NARA_API_KEY=입찰공고_서비스키
NARA_PRESPEC_API_KEY=사전규격_서비스키

The .env file is automatically loaded when running the server.

Claude Desktop Configuration

설정 파일 위치:

MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "nara-jangteo": {
      "command": "uvx",
      "args": [
        "--python",
        "3.11",
        "--from",
        "nara-mcp-server",
        "nara-server"
      ],
      "env": {
        "NARA_API_KEY": "입찰공고 API KEY",
        "NARA_PRESPEC_API_KEY": "사전규격 API KEY",
        "UV_LINK_MODE": "copy"
      }
    }
  }
}

중요 사항:

PyPI 설치 시 nara-server 명령어가 자동으로 등록됩니다
NARA_API_KEY — 입찰공고 ServiceKey 입력
NARA_PRESPEC_API_KEY — 사전규격 ServiceKey 입력 (미설정 시 사전규격 검색 비활성화)
Claude Desktop 재시작 필요

Other MCP Clients

Continue, Cline 등 다른 MCP 클라이언트도 동일한 방식으로 설정 가능합니다.

Development & Testing

Using .env File (Recommended)

For local development, create a .env file:

# .env
NARA_API_KEY=입찰공고_서비스키
NARA_PRESPEC_API_KEY=사전규격_서비스키

The .env file is automatically loaded and not tracked by git (.gitignore).

Example workflow:

# 1. .env 파일 생성 후 두 API 키 입력
# NARA_API_KEY=your_bid_key
# NARA_PRESPEC_API_KEY=your_prespec_key

# 2. Run MCP Inspector (no env setup needed!)
npx @modelcontextprotocol/inspector uv --directory . run python -m nara_server.server

Available Tools

1. `get_bids_by_keyword`

키워드로 용역 입찰공고 및 사전규격을 검색합니다. (최대 20개 결과)

파라미터:

keyword (필수): 검색 키워드 (예: "인공지능", "AI", "플랫폼", "시스템 구축")
days (선택, 기본값: 7): 검색 기간 (일). 오래된 공고 검색 시 늘릴 것 (예: 30, 60, 90)

반환 정보:

일반 입찰공고:

공고명 (bidNtceNm)
공고번호 (bidNtceNo)
수요기관 (dminsttNm)
예산 (bdgtAmt / presmptPrce)
마감일시 (bidClseDt) + 진행중/마감 상태 표시
제안요청서 파일 (제안요청서/제안 키워드 포함 파일만 자동 필터링)

사전규격:

품명/서비스명 (prdctClsfcNoNm)
사전규격번호 (bfSpecRgstNo)
발주기관 (orderInsttNm)
배정예산 (asignBdgtAmt)
의견마감일시 (opninRgstClseDt)
규격문서 파일 URL (specDocFileUrl1~5)

예시 질문:

나라장터에서 "인공지능" 키워드로 입찰공고를 검색해줘

AI 관련 정부 프로젝트 입찰 공고를 최근 30일 기준으로 찾아줘

2. `recommend_bids_for_dept`

부서/팀 프로필을 기반으로 맞춤형 입찰공고를 추천합니다.

파라미터:

keyword (필수): 검색 키워드
department_profile (필수): 부서/팀 설명 (예: "UI/UX 디자인팀", "AI/ML 개발팀")
days (선택, 기본값: 7): 검색 기간 (일). 오래된 공고 검색 시 늘릴 것 (예: 30, 60, 90)

검색 범위:

최대 60개 결과 (일반 입찰 30개 + 사전규격 30개)

출력 방식:

사용자가 "Top 5" 또는 특정 개수를 요청하면 해당 개수만큼 추천
"모든 관련 공고"를 요청하면 전체 목록을 적합도 순으로 표시
예산이 있는 항목 우선 추천
제안요청서/과업지시서 파일만 자동 필터링하여 표시

예시 질문:

우리 팀은 클라우드 인프라 구축 전문팀이야. "클라우드" 키워드로 우리 팀에 맞는 입찰공고 Top 5를 추천해줘

데이터베이스 마이그레이션 전문가인데, "DB" 키워드로 관련된 모든 공고를 보여줘

3. `analyze_bid_detail`

입찰공고 첨부파일(제안요청서)을 다운로드하고 텍스트를 추출하여 분석합니다.

파라미터:

file_url (필수): 첨부파일 URL (검색 결과의 제안요청서 URL)
filename (필수): 파일명 (검색 결과의 파일명)
department_profile (선택): 부서 설명 (입력 시 전략 분석 포함)

지원 형식:

HWP: 한글 문서 (주요 형식, langchain-teddynote HWPLoader 사용)
HWPX: 한글 오피스 XML 문서
PDF: 텍스트 기반 PDF (이미지 기반 PDF는 제외)
DOCX: MS Word 문서
XLSX: Excel 스프레드시트
ZIP: 자동으로 내부 파일 선택
- 우선순위: 제안요청서 > 과업지시서 > .hwp/.hwpx > .docx/.pdf

예시 질문:

위 공고의 첨부파일을 분석해줘. 우리 팀은 AI 개발팀이야.

분석 결과:

Fit Score (0-100): 팀과 프로젝트의 적합도
Core Tasks: 팀이 수행할 핵심 업무
Winning Strategy: 입찰 전략 3가지
Risk Factors: 위험 요소 (기술스택, 일정, 페널티 등)

Usage Examples

기본 검색

Q: 나라장터에서 "시스템 개발" 키워드로 입찰공고를 검색해줘

A: 🔍 **일반 입찰 공고 (Regular Bids)**
   Found 15 bid notice(s) total, 15 retrieved
   📅 Search period: 20260116 ~ 20260123

   ## 1. AI 기반 고객관리 시스템 개발 용역
      📌 공고번호: 20260112345-00
      🏢 수요기관: 서울시청
      💰 예산: 150,000,000원
      ⏰ 마감일시: 202601201430 (✅ 진행중)
      📎 제안요청서:
         - 제안요청서_AI시스템.hwp: [URL]

   ================================================================================
   📋 **사전규격 공고 (Preliminary Specifications)**
   Found 3 pre-spec(s) total, 3 retrieved

   ## 1. 📋 [사전규격] 고객관리 플랫폼 구축
      📌 사전규격번호: PRE20260101-01
      🏢 발주기관: 경기도청
      💰 배정예산: 80,000,000원
      ⏰ 의견마감일시: 202601251700 (🔴 마감)
      📎 규격문서:
         - 규격문서 1: [URL]

맞춤형 추천

Q: 우리 팀은 React 기반 웹 프론트엔드 개발 전문팀이야. "플랫폼 구축" 키워드로 우리 팀에 맞는 입찰공고 Top 5를 추천해줘

A: 🎯 Department-Filtered Integrated Search Results

   📋 Department Profile: React 기반 웹 프론트엔드 개발 전문팀
   🔍 Keyword: 플랫폼 구축
   📊 Results:
     - Regular Bids: 22 open (out of 30 total)
     - Pre-Specs: 8 open (out of 30 total)

   **Top 5 Recommendations:**

   1. ✅ [BID] 시민참여 플랫폼 구축 (적합도 95점)
      - 예산: 200,000,000원
      - 이유: React/TypeScript 기반 웹 프론트엔드 구축 명시, UI/UX 디자인 역량 중요

   2. ✅ [PRESPEC] 공공서비스 웹포털 사전규격 (적합도 90점)
      - 예산: 150,000,000원
      - 이유: 반응형 웹 디자인 요구, 사전 의견 제출로 경쟁력 확보 가능
   ...

전략 분석

Q: 위 1번 공고의 첨부파일을 분석해줘

A: 📄 Bid Document Analysis

   📎 File: 제안요청서_시민참여플랫폼.hwp

   ## Strategic Analysis

   **Fit Score:** 92/100
   - React 18, TypeScript, Tailwind CSS 기술스택 요구
   - 반응형 웹 디자인 및 접근성(WCAG 2.1) 준수 필요

   **Core Tasks:**
   1. React 기반 SPA(Single Page Application) 개발
   2. 사용자 대시보드 및 관리자 콘솔 UI 구현
   3. REST API 연동 및 상태 관리 (Redux/Zustand)

   **Winning Strategy:**
   1. 포트폴리오에서 정부기관 반응형 웹 사례 강조
   2. 접근성 준수 경험 및 웹 표준 인증서 제시
   3. React 성능 최적화 기법 (Code Splitting, Lazy Loading) 강조

   **Risk Factors:**
   - ⚠️ 개발 기간 3개월로 촉박함 (일반적으로 4-5개월 소요)
   - ⚠️ 지체상금: 일 0.1% (최대 10%)
   - ✅ 기술스택은 팀 역량과 100% 일치

Troubleshooting

1. ValueError: NARA_API_KEY / NARA_PRESPEC_API_KEY not found

원인: API 키가 환경변수로 설정되지 않았습니다.

해결 방법:

Claude Desktop 설정 파일의 env 섹션에 NARA_API_KEY와 NARA_PRESPEC_API_KEY 추가
사전규격 키(NARA_PRESPEC_API_KEY) 미설정 시 사전규격 검색은 비활성화되지만 일반 입찰공고 검색은 계속 동작합니다
Claude Desktop 재시작

2. No Results Found

원인: 해당 기간에 해당 키워드의 공고가 없습니다.

해결 방법:

다른 키워드로 검색 시도 (더 일반적인 키워드 사용)
days 파라미터를 늘려 검색 기간 확장 (예: days=30, days=60)
마감된 공고는 🔴 마감 표시와 함께 그대로 표시되므로 별도 확인 불필요

3. API Error (Code: 20 - Access Denied)

원인: API 키가 잘못되었거나 활용신청이 승인되지 않았습니다.

해결 방법:

공공데이터포털 > 마이페이지에서 ServiceKey 확인
활용신청 승인 여부 확인

4. HWP 파일 추출 실패

원인:

DRM/암호화된 HWP 파일
비표준 인코딩 또는 손상된 파일
특수한 압축 방식 사용

해결 방법:

원본 링크에서 수동 다운로드 시도
PDF 버전 파일이 있는지 확인
다른 첨부파일(DOCX, PDF 등) 사용

참고:

이 서버는 langchain-teddynote HWPLoader를 사용하여 대부분의 HWP 파일 처리 가능
추출 실패 시 olefile 파서로 자동 폴백

API Information

데이터 출처: 조달청 나라장터 (Korea Public Procurement Service)
엔드포인트:
- 일반 입찰: http://apis.data.go.kr/1230000/ad/BidPublicInfoService/getBidPblancListInfoServcPPSSrch
- 사전규격: http://apis.data.go.kr/1230000/ao/HrcspSsstndrdInfoService/getPublicPrcureThngInfoServcPPSSrch
API 키: 두 서비스 각각 공공데이터포털 활용신청 필요 (NARA_API_KEY, NARA_PRESPEC_API_KEY)
공고 유형: 용역 (Service) - 컨설팅, 개발, SI 프로젝트
검색 기간: 기본 7일, days 파라미터로 자유 설정 가능 (예: 30, 60, 90일)
날짜 청크 분할: API 날짜 범위 제한(~15일) 자동 우회, 내부적으로 15일 단위 분할 후 결과 병합
필터링:
- 모든 공고 표시, 마감일시 옆에 ✅ 진행중 / 🔴 마감 상태 표시
- 사전규격 항목은 📋 [사전규격] 태그로 일반 입찰공고와 명확히 구분

참고:

물품 공고: 엔드포인트 변경 필요 (getBidPblancListInfoThngPPSSrch)
공사 공고: 엔드포인트 변경 필요 (getBidPblancListInfoCnstwkPPSSrch)

Technical Stack

Python: 3.10+
MCP Framework:
- mcp>=1.15.0 - Model Context Protocol SDK
HTTP Client: httpx>=0.27.0 - Async HTTP requests
File Extraction:
- langchain-teddynote>=0.3.9 - Enhanced HWP extraction (primary, with zlib compression support)
- olefile>=0.47 - HWP fallback (legacy MS OLE format parser)
- pypdf>=4.0 - PDF text extraction
- python-docx>=1.1 - DOCX parsing
- openpyxl>=3.1 - XLSX reading
LLM Integration:
- langchain>=0.1.0,<1.0.0 - Document loading framework
- langchain-core>=0.1.0,<1.0.0 - Core LangChain utilities
Utilities:
- python-dotenv>=1.0.0 - Environment variable management

Project Structure

narajangteo_mcp_server/
├── src/
│   └── nara_server/
│       ├── __init__.py          # Package initialization
│       ├── server.py             # Main MCP server (STDIO transport)
│       └── file_extractor.py     # Multi-format file text extraction
├── pyproject.toml                # Python project metadata & dependencies
├── .env                          # Environment variables (local)
├── README.md                     # This file
├── CLAUDE.md                     # Developer guide
└── LICENSE                       # MIT License

Development

Setting up development environment

# Clone repository
git clone https://github.com/Datajang/narajangteo_mcp_server.git
cd narajangteo_mcp_server

# Install in editable mode
pip install -e .

# Set environment variables
export NARA_API_KEY="your_bid_key"          # MacOS/Linux
export NARA_PRESPEC_API_KEY="your_prespec_key"

set NARA_API_KEY=your_bid_key               # Windows
set NARA_PRESPEC_API_KEY=your_prespec_key

# Or use .env file (recommended)
echo "NARA_API_KEY=your_bid_key" > .env
echo "NARA_PRESPEC_API_KEY=your_prespec_key" >> .env

Running the server

# Run directly
nara-server

# Or using Python module
python -m nara_server.server

Testing with MCP Inspector

MCP Inspector provides an interactive UI for testing tools:

# Install Inspector
npm install -g @modelcontextprotocol/inspector

# Run with Inspector (automatically loads .env)
npx @modelcontextprotocol/inspector python -m nara_server.server

The Inspector will open http://localhost:6274 with:

Interactive tool testing
Real-time request/response logs
Tool parameter validation

Prerequisites:

Node.js 18+ (Download)

Contributing

Contributions are welcome! Please follow these guidelines:

Fork the repository
Create a feature branch
Make your changes with clear commit messages
Test thoroughly with real API calls
Submit a Pull Request

License

MIT License - see LICENSE file for details

Author

Datajang (GitHub)

Acknowledgments

조달청 나라장터 for providing the public API
Anthropic for the MCP protocol
Korean government for open data initiatives

Available Tools

3 tools

analyze_bid_detailA

    Download and extract text from bid attachment (RFP/제안요청서) for strategic analysis.
    Supports HWP, HWPX, PDF, DOCX, XLSX, and ZIP files.
    ZIP files are processed with priority: 제안요청서 > 과업지시서 > .hwp > .pdf

    Args:
        file_url: Attachment URL (ntceSpecDocUrl1 from search results)
        filename: Filename (ntceSpecFileNm1 from search results)
        department_profile: Optional - Your team description for strategic analysis.
                           If provided, response includes analysis prompts for Fit Score,
                           Core Tasks, Winning Strategy, and Risk Factors.

    Returns:
        Extracted document text with optional analysis prompts

ParametersJSON Schema

Name	Required	Description	Default
`file_url`	Yes
`filename`	Yes
`department_profile`	No

Output Schema

ParametersJSON Schema

Name	Required	Description
`result`	Yes

TDQS

A4.2/5.0

Behavior4/5

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

Discloses specific file format support (HWP, HWPX, etc.) and ZIP extraction priority logic (제안요청서 > 과업지시서 > .hwp > .pdf) not inferable from schema; explains conditional analysis prompts triggered by department_profile.

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?

Well-structured with Args/Returns sections; front-loaded with core purpose; file priority logic and parameter mappings are high-value details that earn their place; could be slightly more compact but appropriately detailed.

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?

Sufficient for medium complexity: covers file handling behaviors, optional analysis triggers, and references output schema existence; minor gap regarding error handling for unsupported formats or large files.

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

Parameters5/5

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

Excellent compensation for 0% schema description coverage: maps file_url and filename to specific API response fields (ntceSpecDocUrl1, ntceSpecFileNm1) and clarifies department_profile's effect on output (generates analysis prompts).

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?

Clearly states the tool downloads/extracts text from bid attachments (RFPs) for strategic analysis, distinguishing it from sibling search/recommendation tools by focusing on document analysis rather than bid discovery.

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?

Implies workflow context by referencing 'ntceSpecDocUrl1 from search results' in parameter descriptions, but lacks explicit when-to-use guidance contrasting it with get_bids_by_keyword or recommend_bids_for_dept.

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

get_bids_by_keywordA

    Search Korean government procurement notices (나라장터).
    Returns BOTH regular bid notices (입찰공고) AND preliminary specifications (사전규격)
    for service-type (용역) projects including consulting, development, and SI.

    Args:
        keyword: Search keyword for bid title (공고명).
                 Examples: '인공지능', 'AI', '플랫폼', '시스템 구축', etc.
        days: Search window in days from today (default: 7).
              Increase for older bids, e.g. 30, 60, 90.

    Returns:
        Formatted string with bid information

ParametersJSON Schema

Name	Required	Description	Default
`keyword`	Yes
`days`	No

Output Schema

ParametersJSON Schema

Name	Required	Description
`result`	Yes

TDQS

A4.5/5.0

Behavior4/5

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

Since no annotations provided, description carries full burden; it successfully discloses return type (formatted string), scope limitation (service-type/용역 projects only), and dual result types (bid notices + preliminary specifications).

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?

Uses efficient docstring structure (Args/Returns), front-loaded with clear purpose statement, every sentence adds value including Korean translations and specific project examples, appropriate length for complexity.

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?

Given simple 2-parameter schema with output schema available, description adequately covers tool scope, parameter semantics, and return format; mentions important service-type constraint but could note behavior when no bids found.

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

Parameters5/5

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

Schema has 0% description coverage, so description provides essential compensation: explains keyword maps to bid title (공고명) with 4 examples, and days is search window from today with default value context.

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?

Clear specific verb (Search) + resource (Korean government procurement notices/나라장터), and explicitly distinguishes from siblings by stating it returns BOTH regular bid notices AND preliminary specifications for service-type projects only.

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?

Provides concrete examples for keyword parameter ('인공지능', 'AI', etc.) and explicit guidance on days parameter usage (increase for older bids: 30, 60, 90), though lacks explicit comparison to sibling tools analyze_bid_detail and recommend_bids_for_dept.

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

recommend_bids_for_deptA

    Search government procurement notices with department context for personalized recommendations.
    Returns up to 60 results (30 regular bids + 30 pre-specs) with analysis instructions.
    LLM can flexibly present Top N items or all relevant items based on user's request.
    Prioritizes items with non-zero budgets.

    Args:
        keyword: Search keyword (e.g., 'AI', 'Cloud', '플랫폼')
        department_profile: Description of your team/department.
                           Examples: 'UI/UX 디자인팀', 'Database Migration Unit',
                                    'AI/ML 개발팀', '클라우드 인프라팀'
        days: Search window in days from today (default: 7).
              Increase for older bids, e.g. 30, 60, 90.

    Returns:
        Formatted recommendations with strategic analysis

ParametersJSON Schema

Name	Required	Description	Default
`keyword`	Yes
`department_profile`	Yes
`days`	No

Output Schema

ParametersJSON Schema

Name	Required	Description
`result`	Yes

TDQS

A4/5.0

Behavior4/5

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

Discloses result limits (60 items max), budget prioritization logic, and flexible presentation options, compensating for missing annotations.

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?

Well-structured with Args/Returns sections, though the sentence about LLM flexibility ('LLM can flexibly present...') is slightly 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?

Comprehensive for the domain, including department profile examples and output format description, though it could briefly clarify what 'pre-specs' means.

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

Parameters5/5

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

With 0% schema description coverage, the Args section provides crucial examples for all parameters (e.g., 'AI/ML 개발팀', 'Database Migration Unit') and usage guidance for the days parameter.

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?

Clearly states the tool searches procurement notices with department-specific personalization, distinguishing it from generic keyword search (get_bids_by_keyword) and single-bid analysis (analyze_bid_detail).

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?

Provides behavioral guidance (prioritizes non-zero budgets, returns 30+30 results) but lacks explicit guidance on when to use this versus sibling tools.

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

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

3 tool updatesv1.3.2Apr 3, 2026
- First observedanalyze_bid_detail
- First observedget_bids_by_keyword
- First observedrecommend_bids_for_dept

Deploy Server

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

TDQS

A4.2/5.0

Scored 2026-04-18 06:46 across 3 tools

Disambiguation5/5

Each tool has a clearly distinct purpose with no overlap. analyze_bid_detail focuses on document extraction and analysis, get_bids_by_keyword performs general keyword searches, and recommend_bids_for_dept provides personalized recommendations based on department context. The boundaries are well-defined and unambiguous.

Naming Consistency5/5

All three tools follow a consistent verb_noun pattern with clear, descriptive names: analyze_bid_detail, get_bids_by_keyword, and recommend_bids_for_dept. The naming convention is uniform throughout the set, making it easy to understand each tool's function at a glance.

Tool Count4/5

Three tools is appropriate for this government procurement domain, covering search, analysis, and recommendation workflows. While slightly minimal, each tool serves a distinct and valuable purpose, and the count aligns well with the server's focused scope on Korean procurement notices.

Completeness4/5

The toolset covers core workflows effectively: searching bids, analyzing bid details, and providing personalized recommendations. Minor gaps exist, such as the lack of tools for bid submission or tracking, but the available tools enable agents to perform essential procurement analysis tasks without significant dead ends.

Maintenance

ActivityInactive

ResponsivenessNo issues

Resources

Need Help?

Related Servers

Related MCP Connectors

Biz PlannerOAuth
com.biz-planner
AI-powered financial simulator for entrepreneurs — simulate, analyze, and plan any business
Finance
2
painspotter A
ai.painspotter
AI-analyzed startup opportunities from Reddit, Hacker News & Product Hunt, with commercial scores.
Research & Data AI & Machine Learning Search
72
Preuve AI Market ResearchOAuth
ai.preuve
Market research and competitive intelligence for startup ideas - every finding source-linked.
Research & Data Search Web Scraping
1
NowenOAuth
ai.nowen
Research companies with an AI advisor, run vendor comparisons, and generate ranked business ideas.
Product Management Project Management
1

Related MCP Servers

Startup Helper MCP
Search Government Data Open Data
re171113-byte
F
license
B
quality
Not graded
maintenance
Provides comprehensive startup consulting services including commercial area analysis, competitor research, policy funding recommendations, startup checklists, and business trends using Kakao Map API and public data.
Updated 9 months ago (2025-12-28 14:04 UTC)
5
-
Enterprise Operation Insight
Research & Data Finance Search
handaas
F
license
Not graded
quality
D
maintenance
Provides comprehensive data insights into enterprise operations, including company trends, financing history, market rankings, and sentiment analysis. It enables users to perform keyword-based searches and evaluate business scale, tax qualifications, and competitive project profiles.
Updated a year ago (2025-08-12 08:42 UTC)
3
-
MCP Research Server
Research & Data Search
eaglelandsonce
F
license
Not graded
quality
D
maintenance
Provides tools for automated company research, competitor identification, and business model analysis to generate comprehensive business intelligence. It enables users to extract market keywords and synthesize competitive insights via AI-powered research capabilities.
Updated 7 months ago (2026-03-07 18:23 UTC)
-
korean-keyword-mcp
Marketing Research & Data
Whitening-Sinabro
A
license
A
quality
D
maintenance
Enables Korean keyword niche analysis using Naver SearchAd API data for CPC, competition, and search volume, with tools for expansion, scoring, and trend discovery.
Updated 7 months ago (2026-02-21 12:43 UTC)
7
10 npm
MIT

Nara MCP Server (나라장터 입찰공고 검색)

Features

Quick Start

Prerequisites

1. API 키 발급 (필수 — 2가지)

2. Python 환경

Installation

Option 1: From PyPI (Recommended)

Option 2: From Source (For Development)

Configuration

Using .env File (Recommended for Development)

Claude Desktop Configuration

Other MCP Clients

Development & Testing

Using .env File (Recommended)

Available Tools

1. get_bids_by_keyword

2. recommend_bids_for_dept

3. analyze_bid_detail

Usage Examples

기본 검색

맞춤형 추천

전략 분석

Troubleshooting

1. ValueError: NARA_API_KEY / NARA_PRESPEC_API_KEY not found

2. No Results Found

3. API Error (Code: 20 - Access Denied)

4. HWP 파일 추출 실패

API Information

Technical Stack

Project Structure

Development

Setting up development environment

Running the server

Testing with MCP Inspector

Contributing

License

Author

Links

Acknowledgments

Available Tools

Output Schema

Output Schema

Output Schema

Tool Schema Changelog

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

1. `get_bids_by_keyword`

2. `recommend_bids_for_dept`

3. `analyze_bid_detail`