This server provides access to South Korea's Health Insurance Review & Assessment Service (HIRA) disease information and statistics, defaulting to Korean medicine (medTp=2) but also supporting Western medicine (medTp=1). It requires a public data portal (data.go.kr) API service key.
Search diseases by name or code (hira_search_disease): Look up disease names (e.g., 요추염좌, 감기) or ICD-style codes (e.g., J00), with support for pagination and filtering by medical type.
Gender & age statistics (hira_disease_gender_age_stats): Retrieve statistics for a disease code broken down by patient gender and age group for a given year.
Hospitalization vs. outpatient statistics (hira_disease_inout_stats): Get statistics split between inpatient (입원) and outpatient (외래) care for a given year.
Medical institution type statistics (hira_disease_institution_type_stats): Retrieve disease statistics categorized by institution type (e.g., clinic, hospital, tertiary hospital) for a given year.
Regional statistics (hira_disease_region_stats): Get disease statistics broken down by the geographic region of the treating medical institution for a given year.
All tools accept a medTp parameter (1 for Western medicine, 2 for Korean medicine), with 2 as the default. API keys can be provided via URL parameters (key, oc, serviceKey) or HTTP headers (x-api-key, Authorization: Bearer).
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@HIRA Disease MCP 요추염좌를 한방 기준으로 검색해줘"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
건강보험심사평가원 질병정보서비스를 Claude custom connector에서 바로 쓰기 위한 원격 MCP 서버입니다.
한의사 대상 강의에서 사용할 수 있도록 기본값은 medTp=2 한방 기준으로 잡았습니다. 의과(양방) 기준으로 조회하려면 medTp=1을 사용하면 됩니다.
Claude custom connector에 넣을 주소는 아래 둘 중 하나입니다.
대부분의 사용자는 아래 주소만 복사해서 Claude custom connector에 넣으면 됩니다.
https://hira-disease-mcp.vercel.app/mcp?oc=YOUR_DATA_GO_KR_SERVICE_KEY여기서 바꾸는 부분은 마지막의 YOUR_DATA_GO_KR_SERVICE_KEY뿐입니다.
공공데이터포털 인증키는 가능하면 Encoding 인증키를 사용하세요. 인증키에 +, /, = 같은 문자가 들어가면 URL에서 깨질 수 있습니다.
강의자가 hiradisease.vercel.app 같은 별도 주소를 제공했다면 그 주소를 그대로 사용하세요.
https://hiradisease.vercel.app/mcp?key=YOUR_DATA_GO_KR_SERVICE_KEY법률 MCP처럼 짧은 파라미터도 지원합니다.
https://hiradisease.vercel.app/mcp?oc=YOUR_DATA_GO_KR_SERVICE_KEY위의 Deploy with Vercel 버튼을 눌러 본인 Vercel에 배포합니다.
배포가 끝나면 Vercel이 이런 주소를 줍니다.
https://your-project-name.vercel.app그 주소 뒤에 /mcp?key=본인_API키를 붙입니다.
https://your-project-name.vercel.app/mcp?key=YOUR_DATA_GO_KR_SERVICE_KEYRelated MCP server: korean-stat-mcp
YOUR_DOMAIN?YOUR_DOMAIN은 직접 입력하는 단어가 아닙니다.
MCP 서버가 배포된 실제 인터넷 주소를 뜻합니다.
Situation | Connector URL |
공개 서버 바로 사용 |
|
강의자가 별도 도메인을 제공 |
|
내 Vercel 프로젝트 사용 |
|
내 커스텀 도메인 사용 |
|
즉, README나 예시에서 보이는 YOUR_DOMAIN은 아래 부분입니다.
https://YOUR_DOMAIN/mcp?key=...
^^^^^^^^^^^
실제 배포 도메인처음 사용하는 사람은 YOUR_DOMAIN을 직접 고민하지 말고, 강의자가 제공한 완성 URL 또는 Vercel 배포 후 나온 URL을 복사하면 됩니다.
질병명/상병코드 검색
질병 성별·연령별 통계 조회
질병 입원·외래별 통계 조회
질병 의료기관 종별 통계 조회
질병 의료기관 지역별 통계 조회
공공데이터포털의 건강보험심사평가원_질병정보서비스는 XML REST API이며, 질병명칭/코드조회와 여러 질병 통계를 제공합니다.
Source: https://www.data.go.kr/data/15119055/openapi.do
한의사만 사용할 수 있는 도구가 아닙니다.
HIRA 질병정보서비스는 의과(양방)·한방 구분값을 제공하므로, 한의사뿐 아니라 의사, 보건의료 연구자, 보험·심사 관련 실무자, 의료 데이터 교육자도 사용할 수 있습니다.
medTp 값은 아래처럼 사용합니다.
medTp | Meaning |
| 의과(양방) |
| 한방 |
이 repo는 한의사 강의용 편의를 위해 기본값만 medTp=2로 둔 것이고, 도구 입력에서 medTp=1을 지정하면 의과 기준으로 조회할 수 있습니다.
이 서버는 아래 형식을 모두 지원합니다.
https://DEPLOYED_DOMAIN/mcp?key=YOUR_DATA_GO_KR_SERVICE_KEY
https://DEPLOYED_DOMAIN/mcp?oc=YOUR_DATA_GO_KR_SERVICE_KEY
https://DEPLOYED_DOMAIN/mcp?serviceKey=YOUR_DATA_GO_KR_SERVICE_KEY고급 사용자는 헤더 방식도 사용할 수 있습니다.
x-api-key: YOUR_DATA_GO_KR_SERVICE_KEY
Authorization: Bearer YOUR_DATA_GO_KR_SERVICE_KEYURL 쿼리에 API 키를 넣으면 브라우저 기록, 프록시 로그, 배포 플랫폼 로그에 남을 수 있습니다. 강의용·개인용 연결에는 편하지만 공개 공유는 하지 마세요.
회원가입 또는 로그인
건강보험심사평가원 질병정보서비스 검색
원하는 Open API 선택
활용신청 클릭
개발단계로 신청
마이페이지 → 개발계정 → 인증키 확인
Claude connector URL의 key= 또는 oc= 뒤에 입력
Claude connector URL에는 가능하면 Encoding 인증키를 넣으세요.
공공데이터포털 안내에 따르면 개발계정은 자동승인이고, 개발계정 트래픽은 10,000건입니다.
hira_search_disease질병명 또는 상병코드로 HIRA 질병 정보를 검색합니다.
Example:
{
"searchText": "요추염좌",
"diseaseType": "SICK_NM",
"medTp": 2,
"numOfRows": 10
}hira_disease_gender_age_stats상병코드 기준 성별·연령별 통계를 조회합니다.
{
"sickCd": "J00",
"year": 2024,
"medTp": 2
}hira_disease_inout_stats상병코드 기준 입원·외래별 통계를 조회합니다.
hira_disease_institution_type_stats상병코드 기준 의료기관 종별 통계를 조회합니다.
hira_disease_region_stats상병코드 기준 의료기관 지역별 통계를 조회합니다.
처음 사용할 때는 한방, 의과만 쓰기보다 medTp 값을 함께 적어주면 더 안전합니다.
공식 명세 기준:
기준 | medTp |
의과(양방) |
|
한방 |
|
HIRA 질병정보 MCP로 요추염좌를 한방 기준, 즉 medTp=2로 검색하고 관련 상병코드를 알려줘.J00 상병코드의 2024년 한방 기준, 즉 medTp=2 성별·연령별 통계를 요약해줘.감기 관련 상병을 한방 기준, 즉 medTp=2로 검색한 뒤 한의원 강의에서 설명하기 좋게 정리해줘.HIRA 질병정보 MCP로 요추염좌 관련 질병을 먼저 검색해줘. 그 다음 적절한 상병코드를 골라서 한방 기준 medTp=2와 의과 기준 medTp=1의 2024년 성별·연령별 통계를 비교해줘.HIRA 질병정보 MCP로 요추염좌 관련 상병코드를 찾고, 한방 기준 medTp=2와 의과 기준 medTp=1에서 입원·외래 통계가 어떻게 다른지 비교해줘. 강의에서 설명하기 쉽게 핵심만 정리해줘.HIRA 질병정보 MCP가 연결되어 있는지 확인하고, 사용할 수 있는 도구 목록과 각 도구로 할 수 있는 일을 짧게 설명해줘.가장 쉬운 방법은 README 상단의 Deploy with Vercel 버튼을 누르는 것입니다.
CLI로 직접 배포하려면:
npm install
npm run build
npx vercelVercel에 배포되면 /mcp가 /api/mcp로 rewrite됩니다.
https://YOUR_PROJECT.vercel.app/mcp?key=YOUR_DATA_GO_KR_SERVICE_KEYClaude connector에는 Vercel 프로젝트 홈 주소가 아니라, 반드시 /mcp?key=...까지 붙인 주소를 넣어야 합니다.
npm install
npm run devHealth check:
curl http://localhost:3000/healthMCP endpoint:
http://localhost:3000/mcp?key=YOUR_DATA_GO_KR_SERVICE_KEYFor Claude Desktop or local MCP clients:
{
"mcpServers": {
"hira-disease": {
"command": "npx",
"args": ["hira-disease-mcp"],
"env": {
"HIRA_SERVICE_KEY": "YOUR_DATA_GO_KR_SERVICE_KEY"
}
}
}
}HIRA changed the disease API base path to diseaseInfoService1 and operation names ending in 1.
This server uses getDissNameCodeList1, getDissByGenderAgeStats1, getDissByHsptlzFrgnStats1, getDissByClassesStats1, and getDissByAreaStats1.
HIRA medTp code: 1 = medicine/western medicine, 2 = Korean medicine.
The default medTp=2 is intended for Korean medicine lecture use cases.
MIT
Get HIRA disease statistics by gender and age. Use for Korean medicine lecture examples after finding a disease code.
| Name | Required | Description | Default |
|---|---|---|---|
| sickCd | Yes | HIRA disease code, for example J00. Use hira_search_disease first if unknown. | |
| year | No | Statistics year, for example 2024. Defaults to last year. | |
| sickType | No | Disease classification type. HIRA examples commonly use 1. | |
| medTp | No | Medical type. Use 2 for Korean medicine/oriental medicine; use 1 for medicine when needed. | |
| pageNo | No | Page number. Defaults to 1. | |
| numOfRows | No | Rows per page. Defaults to 10. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It discloses the purpose but does not mention output format, pagination behavior, data freshness, rate limits, or any other operational traits. The description is minimal and lacks transparency about key behavioral aspects.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two concise sentences, front-loaded with the action, and contains no redundant information. Every word serves a purpose.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (6 parameters, no output schema, no annotations), the description is insufficient. It does not explain the return structure, pagination implications, or any constraints beyond the parameter defaults. A more complete description would include expected output or link to documentation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all parameters adequately. The description adds no further significant meaning beyond the schema; it only provides a high-level use case. A score of 3 reflects the baseline when schema carries the load.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'Get' and the resource 'HIRA disease statistics by gender and age', and provides a specific use case ('Korean medicine lecture examples after finding a disease code'). It distinguishes itself from sibling tools which focus on other dimensions (in/out, institution type, region).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly states when to use the tool ('after finding a disease code') and implies a prerequisite step (search for disease code first, as noted in the parameter schema). While it does not list alternative tools, the context signals clearly differentiate this tool from siblings.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get HIRA disease statistics split by hospitalization and outpatient care.
| Name | Required | Description | Default |
|---|---|---|---|
| sickCd | Yes | HIRA disease code, for example J00. Use hira_search_disease first if unknown. | |
| year | No | Statistics year, for example 2024. Defaults to last year. | |
| sickType | No | Disease classification type. HIRA examples commonly use 1. | |
| medTp | No | Medical type. Use 2 for Korean medicine/oriental medicine; use 1 for medicine when needed. | |
| pageNo | No | Page number. Defaults to 1. | |
| numOfRows | No | Rows per page. Defaults to 10. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, and the description does not disclose behavioral traits such as pagination limits, data freshness, or error handling for invalid codes. The description only states the basic purpose, leaving agents uninformed about important behaviors like output structure or rate limits.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single concise sentence that directly communicates the tool's core function with no extraneous words. It is well-structured and front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (6 parameters, no output schema), the description is minimal. It fails to explain the return format or how the output is structured, which is critical for an agent to use the tool effectively. The lack of output schema makes the description insufficiently complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the baseline is 3. The tool description itself adds no additional meaning beyond what is already in the input schema parameter descriptions. The param details in the schema are sufficient.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Get'), the resource ('HIRA disease statistics'), and the distinguishing feature ('split by hospitalization and outpatient care'). This differentiates it from sibling tools like hira_disease_gender_age_stats which splits by gender/age.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no explicit guidance on when to use this tool versus alternatives. However, the input schema parameter 'sickCd' includes a hint to use hira_search_disease first if unknown, which is helpful but not comprehensive.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get HIRA disease statistics by medical institution class/type.
| Name | Required | Description | Default |
|---|---|---|---|
| sickCd | Yes | HIRA disease code, for example J00. Use hira_search_disease first if unknown. | |
| year | No | Statistics year, for example 2024. Defaults to last year. | |
| sickType | No | Disease classification type. HIRA examples commonly use 1. | |
| medTp | No | Medical type. Use 2 for Korean medicine/oriental medicine; use 1 for medicine when needed. | |
| pageNo | No | Page number. Defaults to 1. | |
| numOfRows | No | Rows per page. Defaults to 10. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description must disclose behavioral traits. It only states 'Get statistics', implying read-only operation, but does not mention pagination, data limits, error handling, authentication, or data freshness. The description is insufficient to inform the agent of operational characteristics.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence with no wasted words. However, it lacks structure or front-loading of critical details. It earns a good score for brevity but could be more informative while remaining concise.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no output schema and six parameters, the description is too minimal. It does not explain what the statistics look like, how institution type is categorized, or how results are structured. Compared to the sibling tools, it provides insufficient context for an agent to understand its full role.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the baseline is 3. The tool description adds only the aggregation context ('by medical institution class/type') but does not provide new meaning beyond the schemas already provide. It meets the minimum for a fully described schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Get'), the resource ('HIRA disease statistics'), and the grouping dimension ('by medical institution class/type'). It differentiates from sibling tools like hira_disease_gender_age_stats or hira_disease_region_stats by specifying the unique aggregation method.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The main description provides no guidance on when to use this tool versus alternatives. It does not mention prerequisites, use cases, or when to avoid it. The only usage hint ('use hira_search_disease first if unknown') is buried in a parameter description, not in the tool-level description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get HIRA disease statistics by medical institution region.
| Name | Required | Description | Default |
|---|---|---|---|
| sickCd | Yes | HIRA disease code, for example J00. Use hira_search_disease first if unknown. | |
| year | No | Statistics year, for example 2024. Defaults to last year. | |
| sickType | No | Disease classification type. HIRA examples commonly use 1. | |
| medTp | No | Medical type. Use 2 for Korean medicine/oriental medicine; use 1 for medicine when needed. | |
| pageNo | No | Page number. Defaults to 1. | |
| numOfRows | No | Rows per page. Defaults to 10. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description must fully disclose behavior, but it only implies a read operation. There is no mention of potential destructiveness, rate limits, authentication needs, or return format, leaving significant gaps for a tool with six parameters.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence, front-loading the key information. It is concise but could include more detail without becoming verbose, such as clarifying the output or usage context.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Despite good schema coverage, the description lacks completeness for a tool with no output schema and six parameters. It does not explain the returned data structure, pagination behavior, or any constraints beyond parameter defaults, leaving the agent underinformed.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, and each parameter has a description in the schema. The tool description adds no additional meaning beyond the schema, so the baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool retrieves HIRA disease statistics filtered by medical institution region, effectively distinguishing it from siblings like hira_disease_gender_age_stats. The verb 'Get' and resource 'HIRA disease statistics' are specific, and the region scope sets it apart.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus alternatives. The description lacks explicit context, exclusions, or prerequisites, relying solely on the tool name for differentiation.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search HIRA disease names and codes. Defaults to Korean medicine data with medTp=2.
| Name | Required | Description | Default |
|---|---|---|---|
| searchText | Yes | Disease name or code to search, for example 감기, 요추염좌, J00. | |
| diseaseType | No | Search field. SICK_NM searches Korean disease name; SICK_CD searches disease code. | SICK_NM |
| sickType | No | Disease classification type. HIRA examples commonly use 1. | |
| medTp | No | Medical type. Use 2 for Korean medicine/oriental medicine; use 1 for medicine when needed. | |
| pageNo | No | Page number. Defaults to 1. | |
| numOfRows | No | Rows per page. Defaults to 10. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the burden of behavioral disclosure. It adds the key default to Korean medicine (medTp=2) beyond what's in the schema, but does not mention other behavioral traits like pagination behavior, rate limits, or that the tool is read-only.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, zero waste. The description is concise and front-loaded with the core purpose, followed by the key default behavior.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (6 parameters, no output schema, no annotations), the description is adequate but not complete. It covers the purpose and a key default, but lacks context about return format, use cases, or relationship to sibling tools. The rich schema partially compensates.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema already provides full descriptions for all 6 parameters (100% coverage). The description adds minimal extra value beyond restating the default for medTp. Baseline 3 is appropriate since schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it searches HIRA disease names and codes. It distinguishes itself from sibling tools which are all statistics-focused, making its purpose distinct and easy to understand.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit guidance on when to use this tool versus the sibling stats tools. The usage is implied (search first, then possibly use stats), but the description does not provide any when-to-use or when-not-to-use context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0hira_disease_gender_age_statshira_disease_inout_statshira_disease_institution_type_statshira_disease_region_statshira_search_diseaseScored across 5 tools
Each tool targets a distinct statistical dimension (gender/age, inpatient/outpatient, institution type, region) plus a search tool, with no overlap in purposes.
All tools follow the consistent pattern 'hira_disease_<specific_dimension>', using snake_case uniformly.
5 tools is well-scoped for a focused domain of disease statistics, covering core breakdowns without unnecessary bloat.
The set covers search and major statistical breakdowns, but lacks a tool for overall aggregated statistics (e.g., total counts) which could be useful.
Look up Korean drug ingredient and product data by HIRA component and product codes via Pilldoc. V…
Korean fact-verification tools for AI agents: business registration, address, DART, apt prices, laws
WHO ICD-10/ICD-11 diagnosis codes. Lookup, search, chapters via official WHO API.
Access Korea’s G2B procurement and Nara Market data for bid notices, awards, contracts, statistics…