hermes-104-hunter
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., "@hermes-104-hunter Search Taipei backend jobs over NT$60k and estimate real applicant counts"
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.
🏹 Hermes 104 Job Hunter MCP
Hermes 專屬 104 人力銀行求職特助 MCP:集合「極輕量 API 探測」、「真實競爭人數數學逆推」與「Hit-in-the-Loop 代填投遞」優勢於一身的下一代求職自動化神兵。
▍為什麼打造這套工具?
現有的 104 MCP 專案往往面臨兩個極端:
純後台治理型:雖然輕量,但缺乏對手實戰競爭情報,且無法協助將信件帶入應徵視窗。
全重裝瀏覽器型:每次簡單查詢都要喚起幾百 MB 的 Chromium 實例,大量肥大 DOM 迅速灌爆 AI 的 Context Window (Token 暴走)。
hermes-104-hunter 將兩者精華融合,並徹底踐行「極致效率」的架構心法:
⚡ 95% 純 API 運作:日常搜尋、條件過濾、職缺詳情與競爭情報完全不開瀏覽器,毫秒級回應,記憶體佔用小於 50MB。
🧮 獨門數學逆推演算法:突破 104 官方「30 人以上」的模糊黑盒子,以百分比最小公倍數與互斥維度約束,精確反推出真實應徵人數(如精準算出 42 人)與對手 Top 10 技能分佈。
🎭 語氣克隆(拒絕 AI 罐頭感):主動調研你 104 帳號儲存的原版自薦信範本,讓 AI 撰寫客製信時完美繼承你的語氣長度與風格。
🛡️ Hit-in-the-Loop (HITL) 安全鐵律:僅在最後下達準備應徵時,才按需喚醒 Playwright 實體視窗代填自薦信並截圖;絕不替人類自動點擊送出,由使用者肉眼驗證後親手點下最後一步。
Related MCP server: Job Agent
▍架構示意圖
┌─────────────────────────────────────────────────────────────────┐
│ Hermes 104 Job Hunter MCP │
├─────────────────────────────────────────────────────────────────┤
│ 【心法層 (Skill & SOP)】 │
│ • SKILL.md (Hermes 專屬技能規範) │
│ • AGENTS.md (五階段狀態機求職 SOP) │
│ • MY_PROFILE.md (求職者技能樹、薪資底線、黑名單紅線) │
├─────────────────────────────────────────────────────────────────┤
│ 【輕量核心層 (95% 場景 - 純 Python + httpx)】 │
│ • 認證:Chrome Cookie 無感提取 / 檔案掛載 / 匿名模式 │
│ • 搜尋:三級 Token 清洗 + 地區代碼自動轉換 + 排除面議/月薪篩選│
│ • 情報:純演算法逆推真實應徵人數與對手學經歷/Top 10 技能 │
├─────────────────────────────────────────────────────────────────┤
│ 【突擊代填層 (5% 場景 - 按需喚醒 Playwright)】 │
│ • 代填投遞:自動開窗、切換指定範本、塞入客製推薦信並截圖 │
│ • HITL 安全邊界:保留實體視窗,絕不越權點擊送出 │
└─────────────────────────────────────────────────────────────────┘▍五大核心工具
工具名稱 | 模式 | 核心亮點 |
| 純 API | 支援中文地名解析(如「台北市」)、排除面議、薪資下限與完全遠端篩選。三級 Token 清洗,防止 Context 暴走。 |
| 純 API | 結構化提取職缺描述、技能需求、加分條件、學經歷門檻與 HR 聯絡人。 |
| 純 API + 數學 | 【獨門絕技】免啟動瀏覽器! 逆推真實應徵人數,輸出對手學經歷與 Top 10 技能分佈,找出突圍盲點。 |
| 瀏覽器 | 讀取 104 帳號儲存的原版推薦信範本,供 AI 學習個人語氣。 |
| 瀏覽器 (HITL) | 自動打開職缺視窗、切換範本、填入自薦信並截圖。堅守安全邊界,絕不自動送出! |
▍安裝與啟動
本專案使用現代 Python 套件管理器 uv 構建。
1. 複製專案與安裝依賴
git clone https://github.com/HCH725/hermes-104-hunter.git
cd hermes-104-hunter
# 安裝 Python 依賴
uv sync
# 若需使用代填應徵功能,安裝 Playwright 瀏覽器核心
uv run playwright install chromium2. 本地驗證測試
uv run pytest -v▍接入設定 (Configuration)
接入 Hermes Agent
在 Hermes Agent 的 config.yaml 或 MCP 配置中加入:
mcp_servers:
hermes-104-hunter:
command: "uv"
args:
- "run"
- "--directory"
- "/Users/hong/workspace/hermes-104-hunter"
- "hermes-104-hunter"
env:
MCP_104_COOKIE_SOURCE: "chrome" # 可選: chrome, file, env, none接入 Claude Desktop
在 claude_desktop_config.json 中配置:
{
"mcpServers": {
"hermes-104-hunter": {
"command": "uv",
"args": [
"run",
"--directory",
"/Users/hong/workspace/hermes-104-hunter",
"hermes-104-hunter"
]
}
}
}▍認證與 Cookie 設定
hermes-104-hunter 支援多種靈活認證方式:
chrome模式(預設):透過browser-cookie3自動讀取本機 Chrome 中 104 的登入 Session,免手動複製貼上。file模式:指定MCP_104_COOKIE_FILE=/path/to/cookies.json讀取匯出的 cookie。env模式:透過環境變數MCP_104_COOKIE="key=value; ..."直接注入。none模式:純公開匿名訪問。搜尋、職缺詳情與對手情報精算 100% 正常運作!
▍隱私與安全邊界 (Security & Privacy Boundaries)
個資防護:
.gitignore預設嚴格排除MY_PROFILE.md、*.profile.md、.env*與*cookie*.json。請安心將個人完整履歷與薪資底線寫入MY_PROFILE.md,絕對不會意外提交至公開 Git 倉庫。
憑證本機限定:
Chrome Cookie 讀取與本機 Session 快取僅限於本地端執行,絕不向外部第三方傳輸或洩漏憑證。
Hit-in-the-Loop 代填安全防護:
應徵代填功能(
job104_prepare_application)僅代為填入自薦信與截圖,絕對不自動點擊送出,保留使用者肉眼審核與最終決策權。
演算法誠實性:
競爭人數逆推僅在「候選解唯一且資料充分」時才標註為精算值 (
inferred);若存在多組解或資料過於粗略,誠實標註為保留區間不確定性 (ambiguous),絕不以推估值充當精確值。
▍授權條款 (License)
本專案依據 MIT 授權條款 開源。
Available Tools
5 toolsjob104_apply_analysisC
【獨門絕技】免啟動瀏覽器!透過 104 內部統計 API 與數學逆推演算法,精算出真實應徵人數(突破 30 人以上黑盒子),並分析對手學歷、年資經歷分佈與 Top 10 具備技能(找出突圍策略關鍵)。
Inputs: job_url_or_id (required). Returns: object
job104_get_my_templatesB
讀取使用者 104 帳號中儲存的所有自薦信/求職信範本,包含範本標題與內文。供 AI 研讀並繼承使用者口吻,撰寫客製化推薦信時徹底消除 AI 罐頭感。
Inputs: job_url_or_id (required). Returns: object
job104_job_detailC
取得 104 職缺詳細資訊。提取職缺描述、技能需求、加分條件、學歷經歷要求、公司福利與 HR 聯絡人。
Inputs: job_url_or_id (required). Returns: object
job104_prepare_applicationA
【Hit-in-the-Loop 代填應徵】以 Playwright 自動化打開真實職缺視窗、切換指定自薦信範本、填入客製推薦信並截圖。安全鐵律:絕不會替使用者按送出按鈕,視窗保留供人類肉眼驗證後手動點擊最後一步。
Inputs: job_url_or_id (required), dry_run, template_title, cover_letter_text. Returns: object
job104_searchB
搜尋 104 人力銀行職缺。支援地點中文轉換、薪資下限過濾、排除面議與完全遠端篩選,已進行三級 Token 清洗與精簡。
Inputs: keyword (required), area, page, order, +4 more. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 5 tools
Search, detail retrieval, applicant-pool analysis, template reading, and form pre-filling are each distinct operations with little true overlap. The only mild risk is confusing job104_apply_analysis (statistical analysis of competitors) with job104_prepare_application (automating form fill), but the descriptions make the split clear.
All tools share the job104_ prefix and mostly follow a verb_noun pattern (search, get_my_templates, prepare_application, apply_analysis). Minor deviations: job104_job_detail is noun-only and redundantly repeats the namespace, and 'apply_analysis' is a less standard verb form.
Five tools form a tight, purposeful pipeline from search to detail to analysis to template selection to form pre-fill; nothing is superfluous. It is slightly thin for a full job-hunting workflow but well matched to the stated scope.
Core flow (find job, inspect job, gauge competition, pick a template, fill application) is covered, but templates are read-only with no create/update/delete, and there is no way to track application history or status. These gaps force agents to work around missing lifecycle operations.
Maintenance
Related MCP Connectors
Tailor resumes, score ATS fit, find hiring leads, and send cold outreach, from your agent.
searchnapply connects your AI agent to 800K+ open jobs taken straight from company career sites, kept up to date as companies add and close them. Search in your own words with the site's filters (role, location, remote, salary, experience, department), read postings in full, and tailor a resume or cover letter from your searchnapply profile. Save a search as a monitor to collect new postings and track what you applied to or skipped. It never invents a job or a salary, and it does not apply for you: you apply on the company's own page. Free plan included, within a daily call limit.
Analyze job listings against your resume, track applications, and generate cover letters.
Search jobs, tailor your resume, write cover letters, and file applications for you.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables interaction with the Boss直聘 recruitment platform to search for jobs and send automated greetings to recruiters. It features automatic QR code login and security verification using Playwright for seamless session management.1MIT
- AlicenseNot gradedqualityDmaintenanceAn AI-powered job search and application assistant that enables multi-source job hunting, OpenAI matching, cover letter generation, and hybrid application automation via Playwright, Chrome CDP, screen OCR, and manual assist.MIT
- AlicenseNot gradedqualityCmaintenanceExposes the entire job-search workflow as 13 chat-driven tools, letting users refresh a ranked feed of live postings from employers' public hiring APIs, view and score job descriptions, tailor and render truthful one-page LaTeX resumes, review skill gaps and screening answers, and log applications. It keeps a human in the loop, since nothing is ever submitted automatically.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to discover postings from watched company boards ranked by keyword fit, tailor and fact-check resumes against a source-cited profile, render ATS-readable PDFs, submit applications through the user's own browser, and track replies and follow-ups. All ranking, claim checking and email classification run as plain code, so LLM tokens are spent only on writing.1Apache 2.0