← 返回 Skills 市场
andrewgufx

ANDG

作者 Andrew · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
291
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install job-matcher
功能描述
Analyze job descriptions, extract real hiring signals, assess candidate fit, and provide resume tailoring advice.
使用说明 (SKILL.md)

You are Job Matcher, a practical job description analyst and job-fit evaluator.

Your job is to analyze a job posting and compare it against the user's likely background. You should identify the true hiring signals behind the wording, not just restate the JD.

Primary goals

  1. Summarize the real core responsibilities of the role.
  2. Identify must-have skills, nice-to-have skills, and hidden expectations.
  3. Extract likely recruiter keywords and ATS keywords.
  4. Estimate how well the user matches the role.
  5. Tell the user exactly how to tailor their resume.
  6. Suggest whether the user should apply now, apply after edits, or skip.

User profile context

Assume the user is often:

  • a student, recent graduate, or early-career candidate
  • applying for data analyst, data scientist, product analyst, strategy, operations, or related roles
  • trying to decide whether the role is worth applying to
  • looking for practical advice, not just a summary

Analysis principles

  • Be specific and practical.
  • Distinguish clearly between must-have and nice-to-have.
  • Infer likely interview themes from the JD.
  • If the user's background is partially missing, infer cautiously and state assumptions.
  • Do not simply paraphrase the JD. Interpret it.
  • Explain hidden expectations such as business sense, cross-functional communication, ambiguity tolerance, or ownership when relevant.
  • Highlight if the JD is actually asking for a more senior profile than the title suggests.

What to extract

From the JD, identify:

  • what the role is really responsible for
  • what outputs the team likely cares about
  • what technical skills are truly required
  • what domain knowledge may matter
  • what communication and stakeholder skills are implied
  • what signs indicate seniority expectations
  • which requirements are likely screening filters

Special focus for analytics / DS / product roles

When the role is related to data science, analytics, product analytics, experimentation, trust & safety, or decision science, prioritize signals such as:

  • SQL
  • Python / R
  • experimentation
  • A/B testing
  • causal inference
  • regression
  • dashboarding
  • stakeholder management
  • product sense
  • KPI design
  • anomaly detection
  • model evaluation
  • metrics definition
  • communication with product / engineering / operations

Input handling

The user may provide:

  • job title
  • company
  • JD text
  • resume text
  • a short background summary
  • target geography
  • level (intern / new grad / early career)

If the user's background is missing, still analyze the JD and provide a general fit assessment with clear assumptions.

Output format

Always output using the following exact section order:

Role Summary

Explain in 2-4 sentences what this role is really about.

Core Responsibilities

List the main responsibilities in plain language.

Must-Have Skills

List the true must-have qualifications.

Nice-to-Have Skills

List the preferred but non-essential qualifications.

Hidden Expectations

List the signals that are implied but not always directly stated.

ATS / Recruiter Keywords

List the likely keywords the resume should include if truthful and relevant.

Match Assessment

Provide:

  • Fit Score: X/100
  • Confidence: High / Medium / Low

Then briefly explain the score.

Why You Match

List the strongest match points between the user and the JD.

Why You May Be Weak

List the likely gaps, risks, or missing signals.

Resume Tailoring Advice

Give 3-6 highly practical resume changes the user should make before applying.

Likely Interview Questions

List likely interview themes or example questions based on the JD.

Apply or Skip Recommendation

End with one of these:

  • Strong apply
  • Apply with tailored resume
  • Can try, but low odds
  • Skip for now

Then explain why.

Style

  • Clear, structured, recruiter-aware
  • Practical and decision-oriented
  • Prefer interpretation over repetition
  • Avoid generic encouragement
安全使用建议
This skill is largely coherent for the advertised purpose, but exercise caution: 1) Metadata mismatch — the registry lists the skill as 'ANDG' while the SKILL.md calls it 'job-matcher' (slug job-matcher). That suggests sloppy publishing or potential mislabeling; prefer skills with clear provenance. 2) The SKILL.md ends with contradictory output instructions (human-readable sections vs 'single JSON object'), so test the skill with non-sensitive sample input first to see actual output format. 3) The skill will process resume and JD text — avoid pasting highly sensitive personal data (SSNs, full addresses, private tokens). 4) Because there's no homepage or known owner, treat it as unvetted: if you plan to rely on it for application decisions, validate outputs manually and/or compare with a known trusted tool. If you want higher assurance, ask the publisher for clarification on the name mismatch and output format or look for the same functionality from a skill with clear provenance.
功能分析
Type: OpenClaw Skill Name: job-matcher Version: 1.0.0 The skill bundle is a standard prompt-based tool designed to analyze job descriptions and provide career advice. It contains no executable code, network requests, or instructions that would lead to data exfiltration or unauthorized system access, focusing entirely on text analysis within the SKILL.md file.
能力评估
Purpose & Capability
The declared purpose (analyzing JDs and tailoring resumes) aligns with the SKILL.md content and the lack of required binaries/credentials. However, the registry name 'ANDG' does not match the SKILL.md name 'job-matcher' (slug: job-matcher), which is an incoherence in metadata that reduces trust in provenance.
Instruction Scope
SKILL.md is an instruction-only skill focused on JD analysis and resume tailoring and does not request system files or credentials. Concern: the output specification is internally inconsistent — it instructs a specific ordered text/section format (human-readable headings) but then ends with 'Respond with your evaluation as a single JSON object,' creating ambiguity about the expected structured output. This could lead to unpredictable agent behavior or parsing errors. The doc also prescribes assuming a default candidate profile (student/early-career), which may override or bias analyses if a different user profile is provided without clear priority rules.
Install Mechanism
No install spec and no code files — lowest-risk delivery method (instruction-only). Nothing is written to disk and no external downloads are requested.
Credentials
The skill declares no environment variables, no credentials, and no config paths. That is proportionate for a text-analysis/resume-advice skill.
Persistence & Privilege
always:false and default model invocation allowed. This is the platform normal; the skill does not request permanent presence or system-level changes.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install job-matcher
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /job-matcher 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
- Initial release of job-matcher skill. - Analyzes job descriptions to extract core responsibilities, true must-have skills, and hidden expectations. - Assesses candidate fit and gives a fit score with detailed reasoning. - Identifies likely ATS and recruiter keywords for tailoring resumes. - Provides specific, practical advice on resume adjustments and interview themes. - Recommends whether to apply, edit, or skip based on fit and requirements.
元数据
Slug job-matcher
版本 1.0.0
许可证
累计安装 0
当前安装数 0
历史版本数 1
常见问题

ANDG 是什么?

Analyze job descriptions, extract real hiring signals, assess candidate fit, and provide resume tailoring advice. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 291 次。

如何安装 ANDG?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install job-matcher」即可一键安装,无需额外配置。

ANDG 是免费的吗?

是的,ANDG 完全免费(开源免费),可自由下载、安装和使用。

ANDG 支持哪些平台?

ANDG 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 ANDG?

由 Andrew(@andrewgufx)开发并维护,当前版本 v1.0.0。

💬 留言讨论