Skip to main content
Glama
mcp-server-104

mcp-server-104

by a7512cs

mcp-server-104

An MCP server for Taiwan's 104 Job Bank (104人力銀行). Lets Claude (or any MCP client) search 104's live job listings directly.

Is this tool right for you?

Your situation

Best tool

Occasionally job hunting on your own

Just open the 104 website

Want to write a one-off crawler to scrape data

A Playwright / cycletls script is enough, no MCP needed

Want Claude to analyze/compare/summarize/automate job listings

This MCP

Related MCP server: LinkedIn MCP Server

Installation

Pick one of the following three, depending on which client you use:

A: Clients with quick-command support — one line, config written automatically:

claude mcp add job104 -- npx -y mcp-server-104   # Claude Code
codex mcp add job104 -- npx -y mcp-server-104    # OpenAI Codex CLI(新版才有;舊版走 B 的 TOML)

B: Clients where you paste config manually — paste the config into that client's MCP config file:

Claude Desktop / Cursor / Windsurf (JSON):

{
  "mcpServers": {
    "job104": { "command": "npx", "args": ["-y", "mcp-server-104"] }
  }
}

OpenAI Codex CLI legacy (~/.codex/config.toml):

[mcp_servers.job104]
command = "npx"
args = ["-y", "mcp-server-104"]

A and B do the same thing: tell the client "start this server with npx". The core is npx -y mcp-server-104 everywhere; the only difference is how each client registers it.

⚠️ ChatGPT web/desktop cannot connect to this kind of local (stdio) server — it only supports remote URL-based MCP, and there's no machine of yours on its cloud to run npx.

C: Developers who want to modify the code — after cloning this repo:

npm install && npm run build
claude mcp add job104 -- node /你的路徑/104-mcp-server/dist/index.js

See "Development" below for daily commands and testing strategy.

How data is fetched

104's search API sits behind Cloudflare bot protection. Using curl or Node fetch (even with Referer / User-Agent headers) gets blocked — returning 403 or Cloudflare's "Just a moment..." challenge page.

The key isn't the header, it's the TLS fingerprint. Cloudflare checks the TLS handshake fingerprint (JA3); a normal program's fingerprint clearly isn't a browser and gets blocked outright.

This project uses cycletls to impersonate Chrome's TLS fingerprint, making Cloudflare think the request comes from a real browser → let through. This way no browser is needed (an order of magnitude lighter, faster, and easier to deploy than Playwright / Selenium), and real JSON is obtained over plain HTTP.

cycletls is backed by a Go-based TLS client subprocess, started once when the server launches and shared for the entire session.

What's available now

Tool

Status

Description

search_jobs

✅ Real data

Search jobs by keyword + multiple filters, with pagination

get_job_detail

✅ Real data

Get full details for a single job: complete JD, salary, location, education/experience requirements, skills, language abilities, benefits, industry

get_company_jobs

✅ Real data

List all open positions at a given company (paginated)

search_jobs parameters

Parameter

Required

Description

keyword

✅

Job title keyword, e.g. Rust 工程師

area

Work location name, e.g. 台北市, 新竹 (auto-resolved to 104's official area codes for querying). Same-name multi-location areas (e.g. "信義區" exists in both Taipei and Keelung) are not searched directly; instead an ambiguousArea candidate list is returned for the model to confirm with you

salaryMin

Minimum monthly salary (TWD), e.g. 60000. Listings clearly below this are filtered out; "negotiable" is kept by default

excludeNegotiable

Set true to exclude "negotiable" listings. Default false

excludeFeatured

Set true to exclude 104's paid ad listings (those with featured=true). Default false

jobCategory

Job category name, e.g. 軟體工程師 (auto-resolved to 104's official job class codes for querying)

remote

Remote: full fully remote / partial partially remote / any either

jobType

Employment type: fulltime full-time / parttime part-time

experience

Years of experience required: under-1y / 1-3y / 3-5y / 5-10y / over-10y

page

Page number (20 per page), default 1. Flip further for more results

limit

Max results returned on this page, up to 20, default 5

Implementation details of filter parameters (all derived from observing 104's official site UI requests + verifying against metadata.total):

  1. salaryMin must be sent together with scmin + sctp=M + scstrict=1; without scstrict the salary filter is completely ignored.

  2. "Negotiable" salary value is 0, and 104 keeps these by default (negotiable could be very high). excludeNegotiable removes them.

  3. The salary cap 9,999,999 is 104's "no upper limit" sentinel value, normalized server-side to "N and above". The salary prefix follows the original s10 type (10=negotiable, 30=hourly, 40=daily, 50=monthly, 60=annual) — part-time jobs are mostly hourly, don't read them as monthly.

  4. remoteWork=1 full/2 partial, ro=1 full-time/2 part-time, jobexp=1/3/5/10/99 (mutually exclusive experience bands).

  5. Area/job class use a tree code table + pruning: when a parent node matches (e.g. "新竹縣市"), the parent code is used rather than expanding into a bunch of child codes — expanding too much makes 104 return 400. For areas with the same name in multiple places (e.g. "信義區"), neither union nor search is performed; ambiguousArea is returned for the model to confirm with the user (unioning geographically unrelated places is meaningless). For job classes with multiple matches, union is kept (searching related job classes together is usually what's wanted).

  6. Ad detection: 104 stuffs ads at the top of results (raw field jobType=1), which ignore the keyword (e.g. a nurse search surfaces "COACH 精品銷售"). Each result carries a featured flag marking it, and excludeFeatured=true filters the whole batch out. jobType=2 (paid priority slots) still matches the keyword and is treated as a valid result without the flag. Search listings deliberately omit the full JD (kept lean, to avoid the model mismatching a listing's URL to another entry when summarizing lists); use get_job_detail for full content.

Field naming is consistent across all three tools (all mapped to 104's original field semantics, avoiding same-name-different-meaning):

Concept

search_jobs

get_job_detail

get_company_jobs

Job code (slug, can be fed back to get_job_detail)

jobId

jobId

jobId

Job URL

url

url

url

Area (district level)

area

area

area

Full address (district + street)

—

location

—

Years of experience required

—

experience

experience

Proficient tools/languages (C++, Linux)

skills

skills

—

Job skills (job-class level, e.g. "軟體工程系統開發")

—

jobSkills

—

Company page URL (feed to get_company_jobs)

companyUrl

companyUrl

—

Whether it's an ad slot (jobType=1)

featured

—

—

Update date (the page's "MM/DD更新")

appearDate

appearDate

—

jobId is always a slug (e.g. 7uqyj), not 104's internal number — only the slug can be fed back to get_job_detail. skills always means "specific technologies". appearDate is uniformly YYYY/MM/DD. Company job listings deliberately omit dates: the company API's raw data only has year-less formats like 8/20, so stale zombie listings always look recently updated (testing found 2025 listings mixed in), silently misleading across year boundaries — if you want a specific listing's date, feed its jobId to get_job_detail for the full version.

get_job_detail parameters

Parameter

Required

Description

jobUrlOrId

✅

Job URL or code, e.g. https://www.104.com.tw/job/7uqyj or 7uqyj (use the url returned by search_jobs)

get_company_jobs parameters

Parameter

Required

Description

companyUrlOrId

✅

Company URL or code, e.g. https://www.104.com.tw/company/1a2x6blghh or 1a2x6blghh

page

Page number (20 per page), default 1

limit

Max results returned on this page, up to 20, default 10

How the three tools chain together:

  • Every result from search_jobs / get_job_detail returns two URLs: url (job) and companyUrl (company).

  • Want the full content of a listing → feed its url to get_job_detail.

  • Want to see "what else this company has open" → feed companyUrl to get_company_jobs (it's a specific company's job list, not a keyword search).

search_jobs ─ url ──────→ get_job_detail
      │                        │
      └─ companyUrl ───────────┴──→ get_company_jobs

104 internal API reference

Main endpoint:

GET https://www.104.com.tw/jobs/search/api/jobs

Required headers: Referer: https://www.104.com.tw/jobs/search/, Accept-Language: zh-TW

Common query parameters (this project currently only uses a subset; the rest are for future expansion):

Parameter

Meaning

Example value

keyword

Keyword

Free text

kwop

Keyword operator

7 (match all)

order

Sort

15 relevance (default) · 16 newest · 13 salary

page / pagesize

Pagination

pagesize recommended 20

area

Area code (comma-separated)

Look up Area.json (see below)

jobcat

Job class code (comma-separated)

Look up JobCat.json

scmin + scstrict=1

Minimum salary

Integer

remoteWork

Remote

1 fully remote · 2 partial · 1,2 either (verified)

ro

Full/part-time

1 full-time · 2 part-time (verified; some wt values return 400, don't use)

jobexp

Experience

1/3/5/10/99 = under 1y/1-3/3-5/5-10/10y+ (mutually exclusive bands, verified)

edu

Education

4,5,6 bachelor's and above, etc.

Area / job class code tables (hosted on static.104.com.tw, no Cloudflare protection, retrievable with a plain fetch):

https://static.104.com.tw/category-tool/json/Area.json
https://static.104.com.tw/category-tool/json/JobCat.json

Other endpoints:

  • Job detail: GET https://www.104.com.tw/job/ajax/content/{slug} (Referer points to /job/{slug})

  • Company jobs: GET https://www.104.com.tw/api/companies/{code}/jobs?page=1&pageSize=20 (returns list.topJobs + list.normalJobs)

File structure

src/
  index.ts            進入點:建 server、掛 tool、接 stdio、處理關閉
  config.ts           所有設定 / 魔術數字(JA3 指紋、endpoint、節流區間…)
  types.ts            乾淨型別 + normalizeJob / JobDetail / CompanyJob(防腐層)
  query.ts            純函式:組查詢網址、client 端過濾、enum 對照
  slug.ts             從 104 網址取出職缺 slug / 公司碼(types/query 共用)
  codes.ts            地區/職類「名稱→官方代碼」解析(樹狀比對+剪枝,快取代碼表)
  api/
    httpClient.ts     cycletls 單例(TLS 指紋偽裝)
    throttle.ts       禮貌性隨機節流 1.5~3.5s
    job104.ts         104 抓取層:組 URL → 打 API → 重試 → 正規化
  tools/
    searchJobs.ts     search_jobs
    getJobDetail.ts   get_job_detail
    getCompanyJobs.ts get_company_jobs
scripts/
  smoke-test.mjs      手動發 JSON-RPC 驗證,不用開 Claude 也能測
test/
  types.test.mjs      normalize 邏輯(薪資格式、面議、哨兵值…)
  query.test.mjs      組網址 / slug / 公司碼 / 過濾 / enum 對照
  codes.test.mjs      代碼表樹狀比對 + 剪枝

Development

npm run build                 # 編譯 src → dist
npm test                      # 跑單元測試(先 build 再 node --test,零額外依賴)
node scripts/smoke-test.mjs   # 煙霧測試(連真實 104)
npm run inspect               # 開 MCP Inspector GUI 除錯

After changing code, run npm run build, then restart Claude Code (or use /mcp reconnect) for changes to take effect — the client only fetches the tool list once at session startup.

Testing strategy: pure logic (normalize, URL building, filtering) is extracted into types.ts / query.ts and tested with Node's built-in node --test — fast and no network needed, so breakage is caught immediately. Network-touching parts (job104.ts / httpClient.ts) are verified against the real 104 via smoke tests.

⚠️ Disclaimer

  • 104 has no public official API. This project uses unofficial internal endpoints from the web frontend, which may stop working at any time due to 104's changes.

  • Automated access may violate 104's terms of service. This project is for personal, low-frequency, learning purposes only.

  • Do not use it for high-frequency scraping, large-scale crawling, or hosting as a public service — you risk being blocked and face legal exposure.

  • This project has built-in polite rate limiting (random 1.5~3.5 second interval between requests); do not remove or lower it.

  • Users are solely responsible for any consequences of using this project.

Available Tools

5 tools
find_company用名稱找 104 公司A

用公司名稱找 104 上的公司,回傳公司名片:companyId(公司代碼,可直接餵給 get_company_jobs)、全名、公司頁網址、產業、地區、員工數、資本額、在徵職缺數。比對是模糊的:英文別名(如 MediaTek)也找得到中文本尊,但 total 含「簡介提及」的公司會偏大。唯一命中或名稱完全相符 → 回單一 company;多家符合 → 回 candidates 候選清單,此時請向使用者確認是哪一家(用產業/地區/在徵職缺數分辨),不要自行猜選。「某公司有沒有某類職缺」的標準流程:find_company 拿 companyId → get_company_jobs 帶 keyword。

Inputs: name (required).

get_apply_analysis取得 104 職缺的應徵者分析A

取得單筆 104 職缺的應徵者分析:兩週內不重複應徵人數(total —— 職缺頁與 search_jobs 的 applyRange 只有區間)與應徵者組成:學歷、年齡、年資、語言(含程度)、科系、技能、證照(每項有 count 人數與 percent 比例)。104 已不提供性別。total 的來源看 totalBasis:api=104 直接給的;inferred=104 自 2026-10 起不再直接給人數,改由它的比例反推、並用人數區間驗證過;unknown=推不出來,total 與 count 為 null,只引用 percent,不要自己估人數。totalCandidates 多於一個(多半是「30 人以上」)時 total 取最小值 —— 只有在每一項人數剛好有公因數時才會是倍數,回答時可註明「推算值」。⚠️ 只在使用者明確要求「這個職缺的應徵者/應徵人數/競爭狀況」時才呼叫;搜尋職缺、看職缺詳情、列公司職缺時不要主動呼叫,也不要為了順便比較競爭度而對多筆職缺批次呼叫。科系/技能/證照只回前 10 名,count 加總不等於 total。資料由 104 每日更新一次(updateTime)。傳入 search_jobs / get_company_jobs 回傳的 url 或 jobId。

Inputs: jobUrlOrId (required).

get_company_jobs列出某公司所有職缺A

列出「某一家指定公司」在 104 上在徵的職缺(職稱、地區、薪資、學經歷要求、網址),可分頁。公司用 find_company 回傳的 companyId,或 search_jobs / get_job_detail 回傳的 companyUrl。帶 keyword 可在這家公司內搜職缺(比對職稱與 JD 內文,含「其他條件」欄 —— 「某公司有沒有 C++」這種問題用它,別自己翻頁過濾職稱)。⚠️ keyword 多字詞是 OR 不是 AND,要同時符合請分次搜再交集。要完整翻頁,limit 請直接用檔位值本身(20、50 或 100)且全程不變 —— 非檔位值的 limit 會截掉每個窗口的尾端且下一頁不補回。置頂職缺(pinned=true)只在第 1 頁回、不佔 limit 名額;total 不含置頂。

Inputs: companyUrlOrId (required), page, limit, keyword.

get_job_detail取得 104 職缺詳情A

取得單筆 104 職缺的完整詳情:完整職務說明、薪資、地點、學經歷要求、需求技能、語言能力、福利、產業別。傳入 search_jobs 結果中的職缺網址(url 欄位)。

Inputs: jobUrlOrId (required).

search_jobs搜尋 104 職缺A

依關鍵字與篩選條件搜尋台灣 104 人力銀行的即時職缺,回傳職稱、公司、地區、薪資、需求技能、兩週內應徵人數區間(applyRange)與職缺網址。支援地區、薪資下限、職類、遠端、全/兼職、年資篩選,以及分頁。若 area 同名多處(如「信義區」有台北市與基隆市兩個),不會直接搜尋,改回傳 ambiguousArea 候選清單 —— 此時請向使用者確認是哪一個,再用完整名稱(如「台北市信義區」)重新搜尋。limit 只數一般職缺:廣告位(featured=true)另計、不佔名額。要完整翻頁請用 limit=20(上游每頁固定 20 筆一般職缺);跨頁彙整時用 jobId 去重(最新排序下新缺插入會使分頁窗口飄移)。keyword 適合職務/技能詞;輸入公司名稱時 104 不執行搜尋,本工具會回 companyKeyword 提示 —— 此時請改用 find_company 找到公司,再用 get_company_jobs 取得該公司職缺。

Inputs: keyword (required), 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

    A4.4/5.0

    Scored across 5 tools

    Disambiguation4/5

    Each tool targets a fairly distinct purpose: search_jobs (keyword search), find_company (company lookup), get_company_jobs (jobs at one company), get_job_detail (single job details), and get_apply_analysis (applicant stats). The pairings find_company/get_company_jobs and search_jobs/get_job_detail could be momentarily confused but the descriptions explicitly disambiguate the workflows and when to call each.

    Naming Consistency4/5

    All names are snake_case verb_noun, which is a clear, readable pattern (search_jobs, get_apply_analysis, find_company, get_company_jobs, get_job_detail). The only minor inconsistency is the mix of 'find_' versus 'get_' for lookup operations, but it remains predictable.

    Tool Count5/5

    Five tools is well-scoped for a job-bank server covering search, single-job detail, company lookup, company job listing, and applicant analysis. Each tool earns its place with no redundancy or filler.

    Completeness4/5

    The read-only lifecycle is well covered: search by keyword, drill into a job, find a company, list its jobs, and analyze applicants. There is no obvious critical gap for a job-search domain, though it lacks any salary/trend or aggregated discovery tools.

    Maintenance

    ActivitySlowing
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      B
      maintenance
      Searches 104 job listings with natural-language filters and retrieves full postings via MCP tools.
      3
      124 PyPI
      32
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      Enables job search on LinkedIn through MCP tools, including keyword and location search, filtering by remote, easy apply, experience level, job type, and date, and retrieving job details.
      -
    • A
      license
      A
      quality
      B
      maintenance
      This MCP server lets AI assistants search job listings on Taiwan's 104 Job Bank, read full job details, and prepare applications with human-in-the-loop confirmation before final submission.
      11
      10
      ISC
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables users to search job postings with keyword, location, remote-only, date, and limit filters, and retrieve individual job details via MCP tools.
      MIT