job-match

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

/job-match：岗位匹配分析

/job-match: Job Match Analysis

把招聘 JD 拆成可核验的要求，再与用户已有证据逐项对照。目标不是生成一个看似精确的匹配分数，而是帮助用户判断：哪些要求已经有证据、哪些只是简历没写清、哪些确实需要补足，以及是否值得投入一次申请。

Breaks down a job JD into verifiable requirements and compares them item by item with the user's existing evidence. The goal is not to generate a seemingly precise match score, but to help users determine: which requirements already have supporting evidence, which are just not clearly stated in the resume, which genuinely need to be filled, and whether it is worth investing in an application.

输入

Input

优先读取当前对话已有材料，不要求用户重复提供：

一个明确的目标 JD；
当前简历、经历说明或主张—证据账本；
可选的求职偏好，例如地点、薪资、岗位级别、行业和不可妥协条件。

没有完整 JD 时，可以基于用户提供的岗位截图、链接或文字片段分析，但必须标明信息边界。只有职位名称、没有岗位要求时，先请求 JD，不根据职位名称猜测具体门槛。

Prioritizes reading existing materials from the current conversation and does not require users to provide them repeatedly:

A clear target JD;
Current resume, experience description, or claim-evidence ledger;
Optional job search preferences, such as location, salary, job level, industry, and non-negotiable conditions.

When a complete JD is not available, analysis can be based on job screenshots, links, or text snippets provided by the user, but the information boundaries must be clearly marked. If only the job title is available without job requirements, request the JD first and do not guess specific thresholds based on the job title alone.

分析方法

Analysis Methodology

从 JD 提取岗位职责、硬性条件、核心能力、加分项和申请约束；合并重复表述，不把公司宣传语当作候选人要求。
为每项要求寻找用户材料中的直接证据，记录证据位置和个人边界。团队成果不能自动视为个人能力。
将匹配状态限定为：
- ```
已匹配
```
  ：存在明确、可展开的事实或作品证据；
- ```
表达缺口
```
  ：用户做过相关工作，但简历或材料没有清楚呈现；
- ```
证据不足
```
  ：材料暗示可能具备，但无法支撑对外表述；
- ```
真实缺口
```
  ：现有事实明确不满足或用户确认尚未具备；
- ```
待确认
```
  ：缺少决定性信息，不能替用户猜测。
先判断硬性条件，再评估核心职责与加分项。地点、学历、工作授权、年限或必备证书等门槛应单独展示，不被其他优势抵消。
给出
```
建议投递
```
、
```
补充材料后投递
```
、
```
谨慎投递
```
或
```
暂不建议投递
```
，并列出促成结论的关键证据。不要输出伪精确的百分比匹配分数。

Extract job responsibilities, hard requirements, core competencies, bonus items, and application constraints from the JD; merge duplicate statements and do not treat company promotional language as candidate requirements.
Find direct evidence in the user's materials for each requirement, and record the location of the evidence and personal boundaries. Team achievements cannot be automatically regarded as personal capabilities.
Limit match statuses to the following:
- ```
Matched
```
  : There is clear, expandable evidence of facts or work products;
- ```
Presentation Gap
```
  : The user has done relevant work, but it is not clearly presented in the resume or materials;
- ```
Insufficient Evidence
```
  : Materials suggest possible possession, but cannot support external statements;
- ```
Real Gap
```
  : Existing facts clearly do not meet the requirement or the user has confirmed that they do not yet possess it;
- ```
To Be Confirmed
```
  : Critical information is missing and cannot be guessed on behalf of the user.
Judge hard requirements first, then evaluate core responsibilities and bonus items. Thresholds such as location, education, work authorization, years of experience, or required certificates should be listed separately and not offset by other advantages.
Provide
```
Recommended to Apply
```
,
```
Apply After Supplementing Materials
```
,
```
Apply with Caution
```
, or
```
Not Recommended to Apply for Now
```
, and list the key evidence supporting the conclusion. Do not output pseudo-precise percentage match scores.

默认交付

Default Deliverables

按以下结构输出：

岗位结论：一句话说明建议和最关键原因；
要求—证据矩阵：JD 要求、重要性、用户证据、匹配状态、风险或下一步；
硬性门槛：逐项列出已满足、未满足和待确认项；
优先补强：最多 5 项，区分“调整表达”“补证据”和“补能力”；
投递策略：是否建议投递、适合使用的简历版本以及需要确认的问题；
后续交接：仅在用户需要时生成对应的下一步提示。

矩阵中的“用户证据”必须能回指简历段落、项目、作品、PR、文件或用户明确陈述。找不到证据时写“未提供”，不能用常识补齐。

Output in the following structure:

Job Conclusion: One sentence stating the recommendation and the most critical reason;
Requirement-Evidence Matrix: JD requirements, importance, user evidence, match status, risks or next steps;
Hard Thresholds: Itemized list of met, unmet, and to-be-confirmed items;
Priority Improvements: Up to 5 items, categorized into "adjust presentation", "supplement evidence", and "build competence";
Application Strategy: Whether application is recommended, suitable resume version to use, and questions that need confirmation;
Subsequent Handoff: Generate corresponding next-step prompts only when the user needs them.

The "user evidence" in the matrix must be traceable to resume paragraphs, projects, work products, PRs, documents, or explicit statements from the user. If no evidence is found, write "Not Provided" and do not fill in with common sense.

路由与交接

Routing and Handoff

需要改写简历 bullet、摘要或 HR 开场白：把矩阵中的
```
表达缺口
```
和已确认事实交给
```
/great-resume
```
；
需要生成或编辑 HTML/PDF 简历：交给
```
/make-resume
```
；
需要预测问题或模拟面试：把高风险要求和
```
证据不足
```
项交给
```
/interview
```
；
已决定申请并需要填写网页：把已确认字段交给
```
/job-apply
```
；
需要记录申请状态：交给
```
/offer
```
。

分析本身不授权修改简历文件、填写申请、发送消息或提交投递。用户只问“匹不匹配”时，到投递建议为止。

Need to rewrite resume bullets, summaries, or HR opening lines: Hand off the
```
Presentation Gap
```
items and confirmed facts in the matrix to
```
/great-resume
```
;
Need to generate or edit HTML/PDF resumes: Hand off to
```
/make-resume
```
;
Need to predict questions or conduct mock interviews: Hand off high-risk requirements and
```
Insufficient Evidence
```
items to
```
/interview
```
;
Have decided to apply and need to fill out web forms: Hand off confirmed fields to
```
/job-apply
```
;
Need to record application status: Hand off to
```
/offer
```
.

The analysis itself does not authorize modifying resume files, filling out applications, sending messages, or submitting applications. When the user only asks "does it match", stop at the application recommendation.

事实边界

Fact Boundaries

不为了贴合 JD 添加用户没有做过的技能、职责、职级、项目或指标；
不把相邻技术自动视为掌握，例如会 React 不等于具备 Vue 项目经验；可以说明可迁移性，但状态仍按证据判断；
不把“优先、熟悉、了解”统一解释为硬性要求，结合 JD 原文和职责上下文判断；
不因单个关键词缺失直接否定岗位，也不因关键词重合就判断胜任；
涉及时效性条件或外部职位页面时，保留抓取日期，页面失效后标记为待重新确认。

Do not add skills, responsibilities, job levels, projects, or metrics that the user has not done just to fit the JD;
Do not automatically regard adjacent technologies as mastered; for example, knowing React does not equal having Vue project experience; transferability can be noted, but the status is still judged based on evidence;
Do not uniformly interpret "preferred", "familiar with", "understand" as hard requirements; judge in combination with the original JD text and responsibility context;
Do not reject a job directly due to the absence of a single keyword, nor judge competence solely based on keyword overlap;
When time-sensitive conditions or external job pages are involved, retain the crawl date, and mark as to be reconfirmed after the page becomes invalid.