Skip to main content
Glama
tonywang0122

買屋快搜

by tonywang0122
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

買屋快搜

房屋物件搜尋 MCP Server — 讓 AI 幫你找房子。

支援 Claude Desktop / Codex CLI，以及所有相容 MCP 的 AI 工具。

一鍵安裝

macOS / Linux：

curl -LsSf https://raw.githubusercontent.com/tonywang0122/house-search-mcp/main/install.sh | bash

Windows (PowerShell)：

irm https://raw.githubusercontent.com/tonywang0122/house-search-mcp/main/install.ps1 | iex

如果遇到 ExecutionPolicy 錯誤：

# 方法 A：只對這次生效（推薦）
powershell -ExecutionPolicy Bypass -Command "irm https://raw.githubusercontent.com/tonywang0122/house-search-mcp/main/install.ps1 | iex"

# 方法 B：永久放行（需管理員權限）
Set-ExecutionPolicy -Scope CurrentUser -ExecutionPolicy RemoteSigned
# 然後重跑安裝指令

安裝腳本自動偵測並設定：

  • Claude Desktop（macOS / Windows 一般版 / Store 版）

  • Codex CLI（macOS / Windows 一般版 / Store 版）

安裝完成後，重啟應用即可使用。

Related MCP server: mcp-tw-lvr

升級

開啟終端機（macOS: Terminal / Windows: PowerShell），貼上：

uv cache clean house-search-mcp

然後重啟 Claude Desktop / Codex CLI，會自動下載最新版。

功能

  • house_search — 搜尋物件列表（城市、行政區、房數、價格、坪數、屋齡等篩選）

  • house_get_detail — 查詢單一物件完整明細（座向、建築結構、生活圈步行距離、管理費等）

使用方式

在 AI 對話中直接說：

  • 「幫我找板橋三房 預算一千八」

  • 「台中西屯有什麼降價的三房大樓嗎」

  • 「查一下這間物件的明細：1083GN」

手動安裝

如果一鍵安裝失敗，可以手動設定：

  1. 安裝 uv

  2. Claude Desktop — 編輯設定檔：

    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

    • Windows: %APPDATA%\Claude\claude_desktop_config.json（Store 版路徑不同，安裝腳本會自動偵測）

    {
  "mcpServers": {
    "買屋快搜": {
      "command": "uvx",
      "args": ["house-search-mcp"]
    }
  }
}

  3. Codex CLI — 執行指令：

    codex mcp add house-search -- uvx house-search-mcp

  4. 重啟應用

Available Tools

2 tools
house_get_detailA

查詢單一房屋物件的完整明細。

回傳所有公開資訊，包含：

  • 基本：名稱、地址、城市、行政區、社區、物件類型

  • 價格：總價、單價、土地單價、原始開價、降價幅度

  • 格局：房/廳/衛/開放式數量、所在樓層、總樓層、屋齡

  • 面積：建物/主建物/土地面積、面積明細、是否有陽台

  • 座向：房屋/大樓/窗戶/土地座向

  • 特徵：邊間、暗房、管理方式、月管理費、車位詳情

  • 建築結構：結構、牆壁、每層戶數、用途、使用分區、注意事項

  • 經紀人賣點描述（description 列表）

  • 標籤：規格/設施/生活機能/特色標籤

  • 媒體：照片 URL 列表、格局圖、3D 格局圖、地圖圖片、VR 看屋連結、 AI 導覽、影片 URL、語音導覽

  • 生活圈：周邊交通/學校/市場/醫療步行距離

  • 經紀人：姓名、電話、門市、照片等完整資訊

  • 其他：經緯度座標、關注人數、首次上架日期、分享連結

Args: house_no: 物件編號（如 '1083GN'），從 house_search 結果的 id 欄位取得。 也可傳入物件 URL，會自動提取編號。

ParametersJSON Schema
NameRequiredDescriptionDefault
house_noYes

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4.5/5.0
Behavior4/5

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

No annotations are provided, so the description must carry the full burden of behavioral disclosure. It explicitly states that it 'returns all public information' and enumerates a comprehensive list of returned categories, making the behavior transparent. It also discloses that a house URL can be passed and the number is auto-extracted, adding input flexibility. This goes beyond a generic 'retrieve details' statement.

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?

The description is lengthy due to the enumerated return categories, but it is well-structured with a clear opening statement and labeled sections. Each category informs the agent about available data, which is valuable for a detail endpoint. It is front-loaded with the primary purpose first, and the Args section is concise.

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

Completeness5/5

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

Given the tool has only one parameter and no annotations, the description provides thorough guidance: parameter source, alternative input, and a comprehensive overview of returned data. The sibling tool house_search is referenced appropriately. There is little ambiguity about what the tool does and how to use it.

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?

The input schema only defines house_no as a required string with no description. The description adds substantial meaning: it specifies the format (e.g., '1083GN'), the source (house_search's id field), and an alternative input type (URL with auto-extraction). This fully compensates for the 0% schema description coverage.

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?

The description opens with a clear statement of purpose: '查詢單一房屋物件的完整明細' (query complete details of a single house object). It specifies a specific verb (query) and resource (single house details), and distinguishes itself from the sibling tool house_search by requiring a single house identifier and referencing house_search as the source for that identifier.

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?

The Args section explicitly instructs that house_no is obtained from house_search results' id field, implying this tool is for post-search detail lookup on a specific property. It does not explicitly state when not to use it, but the reference to house_search provides sufficient contextual guidance for when to invoke this tool versus searching.

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.

  1. 2 tool updatesv0.3.1
    • First observedhouse_get_detail
    • First observedhouse_search
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

A4.6/5.0

Scored across 2 tools

Disambiguation5/5

兩個工具功能明確區分：house_search 用於列表搜尋，house_get_detail 用於單一物件詳細資料。彼此沒有重疊，代理可清楚選擇。

Naming Consistency5/5

兩個工具皆以 house_ 前綴開頭，後接動詞（search、get_detail），命名模式一致且易於預測。

Tool Count4/5

工具數量僅 2 個，略少於典型 3-15 個的範圍，但對於僅需搜尋與查看詳情的房屋查詢服務而言，已足夠且合理。

Completeness5/5

涵蓋房屋查詢的主要流程：先透過 house_search 取得物件列表與篩選條件，再透過 house_get_detail 取得完整細節。此為唯讀服務，沒有明顯缺漏的操作。

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • mcp-tw-lvr

    asgard-ai-platform
    A
    license
    A
    quality
    D
    maintenance
    MCP server for querying Taiwan's real estate transaction registry via web scraping of the Ministry of the Interior's official portal. Enables natural language queries for real estate sales, rentals, and pre-sale housing data.
    1
    1
    MIT

  • Taiwan Tender MCP

    h30190
    A
    license
    Not graded
    quality
    D
    maintenance
    MCP server that provides real-time, high-quality tender information from Taiwan's government e-procurement website via enhanced web scraping, enabling AI to query and analyze procurement opportunities.
    15
    MIT

  • Apartment Hunter MCP Server

    ai-engineers-guild
    A
    license
    A
    quality
    B
    maintenance
    An MCP server for rental apartment aggregation, enabling geographic polygon search, semantic vector search, and LLM-powered analysis. Integrates with AI agents like Claude Desktop to manage apartment hunting workflows.
    15
    2
    MIT