by yanliudesign
Full job-hunt Claude skill bundle — Job Description decoder + Resume builder (11 templates) + Behavioral Interview / story bank. Three self-contained sub-skills with a top-level router.
# Add to your Claude Code skills
git clone https://github.com/yanliudesign/offer-toolkit-skillGuides for using ai agents skills like offer-toolkit-skill.
Last scanned: 7/25/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-07-25T06:18:05.016Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}offer-toolkit-skill is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by yanliudesign. Full job-hunt Claude skill bundle — Job Description decoder + Resume builder (11 templates) + Behavioral Interview / story bank. Three self-contained sub-skills with a top-level router. It has 307 GitHub stars.
Yes. offer-toolkit-skill passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/yanliudesign/offer-toolkit-skill" and add it to your Claude Code skills directory (see the Installation section above). offer-toolkit-skill ships a SKILL.md manifest, so compatible agents can discover and load it automatically.
offer-toolkit-skill is primarily written in HTML. It is open-source under yanliudesign on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh offer-toolkit-skill against similar tools.
No comments yet. Be the first to share your thoughts!
一整套求职 skill 的聚合入口。四条子 skill 各自独立可用,也可以组成完整的求职链路:
种子 JD + 简历 → [Job Hunt List] 搜索 · 去重 · 证据分层 · HTML 清单
↓ 选中岗位
[Job Description Skill] 解码 JD · 出 Offer Strategy Report
↓ 决定投
[Resume Skill] tailor + 美化 · 11 套模板 · 单文件 HTML
↓ 拿到面试
[BQ Skill] 挖故事 · 建故事库 · STAR 化 · 模拟面试
顶层原则(四条子 skill 共用):
| 用户说的话 | 走哪个子 skill |
|---|---|
| "帮我找工作" / "在 LinkedIn 搜适合我的岗位" / "根据这份 JD 找相似职位" / "根据简历推荐岗位" / "整理职位清单" | → job-hunt-list/ |
| 贴出一份 JD / "这个岗位我该不该投" / "帮我看看这份工作" / "match score" / "面试会问什么" | → job-description-skill/ |
| "帮我美化简历" / "改简历" / 上传 PDF/docx / "我没有简历帮我做一份" / "换个模板" / LinkedIn 导入 | → resume-skill/ |
| "帮我准备 behavioral 面试" / "Tell me about a time…" / "帮我挖一个面试故事" / "STAR 怎么写" / "建我的故事库" / "Amazon LP 怎么准备" | → bq-skill/ |
| 一次交出 JD + 简历,想要"全套准备好" | → 先跑 job-description-skill/(生成 Offer Strategy Report,内含 Top 10 面试题) → 用户说"投" → 转 resume-skill/ tailor → "拿到面试" → 转 bq-skill/ JD-driven prep(JD-driven Top 20 + STAR 模板) |
判断不了就问一句:
"你现在在求职链路的哪一步? · 还在找机会,希望自动发现并整理匹配岗位 → Job Hunt List · 看到一个心动岗位,还没决定要不要投 → JD Skill · 已经决定投,需要一份简历 → Resume Skill · 已经拿到面试,要准备行为面试题 → BQ Skill"
每个子目录都是一份完整的 skill(各自有 SKILL.md + 独立 README)。可以聚合装,也可以只装其中一个。
根据简历、目标方向或种子 JD 生成搜索画像,执行多组公开职位查询,采集、去重并区分已核验事实与方向性推断。默认保留所有通过硬条件的唯一岗位,输出可搜索筛选的单文件 HTML 清单;只搜索分析,不自动投递、不处理登录凭据、不绕过访问限制。
把任何 JD 翻译成一份 10 节的 Offer Strategy Report(HTML 单文件,自动打开浏览器):TL;DR 结论 · Match Score · Interview Probability · 公司背景 · JD 深度解码 · JD ↔ 简历逐条比对 · Gap 与补救 · 为什么投/不投 · 薪资 · 下一步 · Top 10 面试题 · 6 周行动计划。
三步:贴 JD → 给简历 → 自动生成 HTML 报告并打开。
把"我需要一份好看的简历"变成稳定流程:所有素材先汇入 schema/resume-data.md 定义的标准数据结构,再套模板渲染。换模板只是换皮,内容不丢。
三条入口:美化已有简历 / LinkedIn 导入 / 对话式建简历。 11 套打印级模板:Classic-ATS · Ledger · Tech Compact · Modern Sidebar · Pillar · Elegant Serif · Atelier · Timeline · Swiss · Executive · Color-block。 每次渲染同时输出锁定版(直接 Cmd+P 存 PDF)和可编辑版(浏览器里点字微调 + 浮动工具条)。
不是背答案,而是建一套可复用的职业故事库。流程:挖掘 → 结构化 → 打标 → 存库 → 复用。
能做的事:
把整个 offer-toolkit-skill/ 目录放进你的 skill 目录(例如 ~/.claude/skills/ 或 VS Code 的 prompts 目录),四条子 skill 会一起被发现。
只想装其中一个? 直接把对应子目录(如 resume-skill/)单独复制过去也可以,每个子 skill 都是自包含的。
MIT — fork, remix, ship your own version.
Created by Dreameryanyan · LinkedIn · X · 小红书
中文 · English
Four job-hunt tools in one agent skill pack — Search · JD · Resume · BQ.
Four job-hunting tools bundled as an agent skill pack. Use the whole thing, or just the piece you need — each sub-folder works on its own.
0 Discover opportunities → job-hunt-list Build a deduplicated, searchable HTML job list.
1 See a job you want → job-description-skill Decode the JD, get an Offer Strategy report.
2 Decide to apply → resume-skill Tailor & polish. 11 print-ready templates.
3 Land the interview → bq-skill Mine stories, build a story bank, prep BQs.
| Sub-skill | What it does |
|---|---|
| job-hunt-list | Give it a resume, target direction, seed JD, or existing job links. It searches public sources, deduplicates candidates, separates verified facts from inference, and generates a searchable HTML job list. It discovers and analyzes; it never applies for you. |
| job-description-skill | Give it a JD and your resume. You get back an HTML report telling you: whether this job is worth applying to, how well you match it, where the gaps are, what you'll probably be asked in interviews, whether the salary is reasonable, and what to do over the next six weeks. |
| resume-skill | Polishes an existing resume, pulls one in from LinkedIn, or builds one from scratch by chatting with you. The content then flows into 11 print-ready templates (Classic-ATS, Ledger, Tech Compact, Modern Sidebar, Pillar, Elegant Serif, Atelier, Timeline, Swiss, Executive, Color-block). Each render gives you two files: one that prints straight to PDF, and one you can click on in the browser to edit. |
| bq-skill | Instead of handing you canned answers, it helps you dig your real past experiences out and organize them into a story bank you can reuse. It asks about your experience step by step, uses STAR/CAR to shape each story, tags them ("took ownership", "handled ambiguity", etc.), and saves them in English and Chinese — so next time you get a different behavioral question, the same story still works. It can also read a JD, predict the 20 questions that company is likely to ask, and walk you through prep for each one. |
The Offer Strategy Report generated by job-description-skill, opened in-browser — verdict, assumptions, TL;DR, and the built-in EN / 中文 toggle in the bottom-right toolbar.
Full galleries: English · 中文 — sub-skill galleries live under each sub-folder's docs/ or assets/.
Drop the whole offer-toolkit-skill/ folder into your skills directory (e.g. ~/.claude/skills/ or VS Code's prompts folder). All four sub-skills get picked up.
Only want one? Copy just that sub-folder — each one is self-contained.
🔒 Your data stays on your machine. JDs, resumes, and stories are read by the model in the moment and written back to your local disk — nothing gets shipped to a server or trained on.
Once installed, just paste any of these into your Claude Code / VS Code chat:
The router picks the right sub-skill. You get an HTML report / resume / story bank on your desktop, ready to open.
| You say | Runs |
|---|---|
| "Find me a job" / "Search LinkedIn for matching roles" / "Build a job hunt list from my resume" | job-hunt-list |
| Paste a JD / "should I apply to this?" / "help me with this job" | job-description-skill |
| "Beautify my resume" / "build me one from scratch" / upload PDF / paste LinkedIn | resume-skill |
| "Prep me for behavioral interviews" / "tell me about a time…" / "mine a story" | bq-skill |
| "Help me find a job" / hand over JD + resume at once | Top-level SKILL.md routes you to the right one |
offer-toolkit-skill/
├── SKILL.md # Router: intent → sub-skill
├── README.md / README.zh.md
├── LICENSE # MIT
├── job-hunt-list/ # ⓪ Complete searchable job discovery list
├── job-description-skill/ # ① JD decoder + Offer Strategy report
├── resume-skill/ # ② Resume builder + 11 templates
└── bq-skill/ # ③ Behavioral interview / story bank
Each sub-folder has its own SKILL.md, README, prompts, frameworks, and templates.
These three also live as separate repos and get maintained there:
This bundle just puts them together with a top-level router.
MIT — fork it, remix it, ship your own version.
Created by Dreameryanyan · LinkedIn · X · Xiaohongshu