job-hunting-mcp
Based on the provided schema, this MCP server offers read-only public job search across 104 and Cake, plus 104-only job detail retrieval.
Search public jobs by keyword, area, job category, salary minimum, job type, remote/new flags, experience, education, sort, page, and page size.
Automatically query enabled 104 and Cake sources without requiring a source input; results include company, title, description, salary, location, source, link, and source-scoped job_ref.
Retrieve full job details using a job_ref returned by search_jobs; the current schema limits get_job_details to 104.
Operate read-only and open-world, without arbitrary URLs, logins, applications, favorites, tracking, or candidate data access.
Note: the README lists compare_jobs and get_company_detail, but they are absent from the provided server schema.
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., "@job-hunting-mcp 幫我找台北的後端工程師職缺,月薪四萬以上,優先看最近更新的"
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.
job-hunting-mcp
🌐 繁體中文 | English
技術版本
job-hunting-mcp 是一個協助個人查詢工作的唯讀 Model Context Protocol(MCP)server。AI agent 可以依自然語言同時搜尋 104 與 Cake 的公開職缺,逐筆顯示資料來源、公司、職稱、工作內容、薪資、地點和原始連結。
Related MCP server: Bossjob Scout MCP
專案範圍
唯讀公開職缺搜尋與詳情查詢
自動查詢 104 與 Cake,不要求使用者輸入來源
每個來源預設 10 筆,可指定 10–20 筆
每筆結果使用來源限定的
job_ref,例如104:8ue17、cake:site-reliability-engineer-sre-71702b不登入、不使用使用者 Cookie、不投遞、不收藏、不追蹤、不讀取求職者資料
安裝
目前必須先從 GitHub 取得專案,再安裝相依套件與建置:
git clone https://github.com/YutangShi/job-hunting-mcp.git
cd job-hunting-mcp
node --version
npm ci
npm run build需求:Node.js 22 LTS、npm,以及可連線至 104/Cake 的網路。impers 使用原生 curl-impersonate 相容 library;若啟動時無法載入,請依錯誤訊息設定原生 runtime。
安裝到 AI agent
Codex
在專案根目錄執行:
NODE_BIN="$(command -v node)"
PROJECT_DIR="$(pwd -P)"
codex mcp add job-hunting-mcp -- "$NODE_BIN" "$PROJECT_DIR/dist/index.js"
codex mcp get job-hunting-mcp重新建置程式後重新啟動 Codex 即可。只有專案路徑或啟動命令改變時才需重新註冊。
Claude Desktop
把以下內容加入 Claude Desktop 的 MCP 設定,並將兩個路徑換成絕對路徑:
{
"mcpServers": {
"job-hunting-mcp": {
"command": "/absolute/path/to/node",
"args": ["/absolute/path/to/job-hunting-mcp/dist/index.js"]
}
}
}macOS/Linux 可用 command -v node,Windows 可用 where node 查詢 Node.js。儲存後重新啟動 Claude Desktop。
ChatGPT 與其他 agent
目前版本是本機 stdio server。Codex CLI 和 Claude Desktop 可以直接啟動本機程序;ChatGPT 網頁版或其他託管型 agent 通常需要遠端可連線的 MCP server、HTTPS endpoint、驗證和該平台的 connector 設定,因此不能直接使用本機 dist/index.js 路徑。這些 client 的遠端部署尚未包含在本版本。
MCP 如何運作
使用者問題
↓
AI agent 呼叫 search_jobs
↓
job-hunting-mcp 分別查詢 104 與 Cake
↓
合併正規化結果,保留每筆來源與來源狀態
↓
AI agent 使用回傳的 job_ref 呼叫 get_job_details本機 client 透過 stdio MCP frame 溝通。來源查詢有節流與大小限制;某個來源失敗時,其他來源的成功結果仍會保留,並在回應中顯示失敗來源與錯誤類別。
MCP tools
目前程式實際註冊 4 個 MCP tools;其中 V1 新增 compare_jobs 與 get_company_detail。smoke:104、smoke:cake、test:e2e 是測試/驗證命令,不是 MCP tools。
Tool | 用途 | 主要輸入 | 回傳內容 |
| 同時搜尋所有啟用來源的公開職缺 |
| 公司、職稱、工作內容描述、薪資、地點、來源、原始連結、 |
| 讀取搜尋結果對應的完整公開職缺內容 |
| 公司、職稱、完整工作內容、條件、薪資、地點、來源、原始連結 |
| 比較 2–10 筆搜尋結果的公司、薪資、地點、條件與資料完整度 |
| 比較狀態、逐筆職缺資料、薪資可比較性、來源、原始連結、失敗項目 |
| 依職缺 |
| 公司名稱、產業、規模、介紹、公司連結、公開職缺、來源、擷取時間 |
search_jobs 不接受 source 欄位;使用者不需要知道職缺來自哪個平台。get_job_details、compare_jobs 與 get_company_detail 都不接受任意 URL,只接受來源限定的 job_ref。get_company_detail 必須先由 search_jobs 建立職缺與公司連結映射。
測試與驗證
npm test
npm run typecheck
npm run build
npm run test:e2e
npm run smoke:104
npm run smoke:cakenpm test:執行 unit、adapter、transport 和 MCP in-memory 測試npm run typecheck:執行 TypeScript 型別檢查npm run build:建立dist/index.jsnpm run test:e2e:啟動建置後的 stdio MCP server,列出全部 4 個 tools,實際呼叫搜尋、詳情、比較與公司詳情;來源被拒絕時驗證明確錯誤狀態npm run smoke:104:即時驗證台北 SRE 多筆搜尋與詳情npm run smoke:cake:即時驗證 Cake SRE 搜尋與詳情;來源回傳 403、429、challenge 或 schema 變更時會明確失敗
授權
本專案採用 MIT License,詳見 LICENSE。第三方地區/職類代碼表的授權說明見 THIRD_PARTY_NOTICES.md。
Available Tools
2 toolsget_job_details讀取職缺詳情ARead-only
讀取職缺完整內容。job_ref 必須來自 search_jobs 結果;目前 V1 只支援來源 104。
Inputs: job_ref (required).
search_jobs搜尋職缺ARead-only
自動搜尋已啟用的 104 與 Cake 公開職缺。直接輸入職務條件即可,不需指定來源;例如 keyword='SRE'。每個來源預設回傳 10 筆,可指定 10–20 筆;來源符合條件的職缺不足時會回傳實際筆數。每筆結果包含公司、職稱、工作內容描述、薪資、原始職缺連結、來源和可供 get_job_details 使用的來源限定 job_ref。
Inputs: edu, area, page, sort, +8 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 2 tools
The two tools are cleanly separated: search_jobs performs discovery across sources, while get_job_details retrieves full content for a specific job_ref produced by search. There is no plausible scenario where an agent would confuse one for the other, and the dependency between them is stated explicitly.
Both names follow the same snake_case verb_noun convention (search_jobs, get_job_details) with a shared '_jobs'/'job' noun stem. The pattern is entirely predictable and would extend naturally to future tools.
Two tools is on the thin side even for a narrowly scoped server; a search-plus-detail pair is the minimum useful surface rather than a well-rounded one. It is defensible for a focused job-search use case but leaves little room for anything beyond the single browse workflow.
Search plus full-detail retrieval covers the core discovery workflow, but get_job_details only supports 104, so Cake results returned by search_jobs are a dead end for detail retrieval. There is also no filtering, sorting, pagination beyond 20 results, or any save/apply/track operation that a 'job hunting' domain would normally imply.
Maintenance
Related MCP Connectors
239k open jobs from 3,519 company career sites on 71 hiring platforms (Workday, Greenhouse, Lever, Ashby and more), with duplicates merged and links to the original posting. Read-only. Free tier without a key: 200 tool calls a day per IP. Send a JobSweep API key as Authorization: Bearer <key> for a paid plan's limits (https://jobsweep.works/pricing).
Career help for people and their AI agents: live jobs from LinkedIn, Indeed and national boards; freelance work, bounties and competitions; scam checks, pay ranges, applying and earnings. Search needs no key.
Search remote jobs, read full postings, and get application links from Claude, Codex, Cursor, and other compatible MCP clients. Tools: search_jobs, get_job, get_filters. Supports cursor pagination and location, category, skill, seniority, employment type, and salary availability filters. Free, read-only Streamable HTTP; no account or API key required. Posting-based facts; remote does not imply worldwide eligibility. Setup: https://heyrole.ai/developers
1.2M+ jobs from employer career sites, salary benchmarks and company hiring signals for AI agents.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to search remote job listings, read full descriptions and structured metadata, compare roles, and retrieve original application links when the account's subscription and scopes allow.29MIT
- AlicenseNot gradedqualityCmaintenanceEnables MCP hosts and agents to search public job listings and retrieve job details via standardized search_jobs and get_job tools, supporting keyword, location, remote, date, and limit filters. It is read-only and rate-limited, returning normalized results with apply and source URLs.MIT
- AlicenseNot gradedqualityCmaintenanceEnables users to search live public job listings, compare roles side by side, retrieve full job details, and render interactive job cards with application links from AI chat clients.MIT
- AlicenseNot gradedqualityBmaintenanceLets AI agents search ranked job listings from company career pages, job boards and the web using plain-language requirements or CV text, with optional location and result limits. It also exposes the available job sources so agents can see where matches come from.2MIT