XBOSS

홈택스 원천세 정기신고 기본정보 저장

hometax__tax__withholding__regular_report__basic_save

Destructive

REQUIRES prior hometax__session__login call with same account_link_id. Save basic taxpayer/report information into a regular-withholding draft. This writes to the institution. The first call is not executed: it returns a preview of exactly what will be sent and a one-time confirmation_token (error WRITE_CONFIRMATION_REQUIRED). Show that preview to the user as it is. Only after the user agrees, call again with the same arguments plus that confirmation_token. Never add the token without the user's agreement. Changing any argument makes the token invalid; call without the token to get a new preview. Exception: with draft_mode=resume the call only reads the return already in progress (nothing is written or deleted at the institution), so it runs at once and returns no confirmation_token. 호출마다 토큰 1소진 요청 처리 시 XBOSS 작업을 생성하고 처리 결과를 저장합니다. 기관 원본의 조회 여부와 별개로 이 호출은 순수 읽기 전용이 아닙니다. 기관 데이터 변경은 도구별 실행 내용과 기존 확인 절차를 따릅니다.

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": {
      "draft_id": "<draft_id>",
      "expected_state_version": "<expected_state_version>",
      "idempotency_key": "<idempotency_key>",
      "basic_document": "<basic_document>"
    },
    "name": "hometax__tax__withholding__regular_report__basic_save"
  }
}

Input Schema

Name	Required	Description
`draft_id`	Yes	Draft ID returned by init
`draft_mode`	No	작성 방식. new(기본값)는 새 신고서로 작성하며, 이미 작성 중인 신고서가 있으면 그것을 지우고 시작합니다. resume 은 작성 중인 신고서를 이어서 씁니다. init 과 basic_save 에 같은 값을 입력해 주세요.
`basic_document`	Yes	Withholding basic document payload
`account_link_id`	No	연동 ID(UUID v4). 선택 입력입니다. 비워 두면 서버가 API 키로 연동을 찾아 씁니다. 같은 기관에 인증서를 여러 개 연동했을 때만 직접 입력해 주세요. 기관마다 ID 가 다릅니다.
`idempotency_key`	Yes	Idempotency key for safe retry
`confirmation_token`	No	첫 호출이 돌려준 1회용 확인 토큰입니다(오류 WRITE_CONFIRMATION_REQUIRED). 같은 인자에 이 값을 그대로 넣어 다시 호출하면 실행됩니다. 10분 뒤 만료되고 한 번만 쓸 수 있습니다.
`expected_state_version`	Yes	Expected draft state version

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 15, 2026

TDQS

A4/5.0

Behavior5/5

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

Annotations only say readOnlyHint=false and destructiveHint=true; the description goes much further, disclosing that this writes to the institution, that the first invocation is a no-op preview with a one-time token, that mutating any argument invalidates the token, and that resume mode performs no write or delete. It also flags per-call token consumption and that a XBOSS job is created, and correctly states the call is not purely read-only, consistent with the annotations.

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

Conciseness3/5

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

The core flow is front-loaded and readable, but the trailing Korean sentences about token exhaustion, XBOSS job creation, and 'not purely read-only' are boilerplate that partially restates the annotations and reads as generic policy text. The prerequisite sentence also precedes the purpose statement, delaying the one-line answer to 'what does this do.'

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, multi-step write tool with no output schema, the description covers prerequisite, confirmation protocol, token lifecycle, and the read-only exception, which is what an agent needs before calling. The remaining gap is the required nested basic_document object, whose structure is left to 'Withholding basic document payload.'

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 schema already documents all seven parameters including token expiry and the new/resume enum meaning, making the baseline a 3. The description adds one genuinely new rule not in the schema - that changing any argument invalidates the confirmation_token - but says nothing about the required nested basic_document payload.

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 states a specific verb and resource: 'Save basic taxpayer/report information into a regular-withholding draft,' which tells an agent exactly what is written and where. It does not explicitly contrast with the 'detail_save' or amendment/late-report 'basic_save' siblings, which is the only thing separating it from a 5.

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?

It gives a hard prerequisite ('REQUIRES prior hometax__session__login call with same account_link_id') and a concrete two-step usage protocol: first call returns a preview and token, second call with the token executes. The draft_mode=resume carve-out is also explained. It stops short of naming when to choose the amendment or late-report variants instead.

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

All XBOSS tools Inspect

Try in Browser

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

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?