agentleFS
kyoungbinkimkyoungbinkim/give-me-job/skills/jd-analyzer/SKILL.md

jd-analyzer

Skill6 starsChanged 5 months agoMITView on GitHub

Analyze a job description, hiring post, or recruitment page for role requirements, evaluation criteria, keywords, risks, and resume or cover letter angles. Use when the user provides a JD URL/text, asks what a company is looking for, wants JD-resume gap analysis, or needs inputs for a tailored cover letter. 한국어 요청도 포함합니다 - 채용공고 분석, 공고 분석, JD 분석, 직무기술서 분석, 자격요건, 우대사항, 이 공고 뭘 원하는지, 공고랑 내 경력 비교.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill to convert a hiring post into practical selection criteria.

    line 8
  2. Distinctive:

    Use this skill when the user provides a JD URL, pasted hiring post, normalized job record, or hiring notes that need Korean-market role …

    line 12
  3. Distinctive:

    Do not use this skill to draft cover letters, invent missing JD requirements, or analyze company culture pages that are not part of the …

    line 16
  4. Distinctive:

    Do not add requirements absent from the JD or role context.

    line 31
  5. Distinctive:

    Do not treat inferred criteria as explicit facts.

    line 32
  6. Distinctive:

    Keep inference conservative.

    line 69
  7. Distinctive:

    Keep eligibility, fit, and preparation effort distinct.

    line 73
  8. Distinctive:

    If the JD is vague, ask for the target role, company, and any available posting text.

    line 109

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

JD Analyzer

Use this skill to convert a hiring post into practical selection criteria.

Trigger

Use this skill when the user provides a JD URL, pasted hiring post, normalized job record, or hiring notes that need Korean-market role analysis.

Do Not Trigger

Do not use this skill to draft cover letters, invent missing JD requirements, or analyze company culture pages that are not part of the hiring post.

Autonomy Level

DoF: MEDIUM

Separate deterministic parsing from conservative inference. Label every inferred evaluation criterion as inferred.

Permitted inferences:

  • Hidden evaluation criteria that follow directly from explicit duties, tools, seniority, or hiring process language.
  • Likely resume evidence categories needed to answer the JD.

Prohibited inferences:

  • Do not add requirements absent from the JD or role context.
  • Do not treat inferred criteria as explicit facts.

Input Contract

Required context:

  • JD URL, JD text, normalized job record, or hiring-post notes.

Optional context:

  • resume.md for gap mapping.
  • Cover letter questions and length limits.

Required parameters:

  • source: URL, text, normalized job record, or unknown.

Outputs produced:

  • applications/<company-role>/jd-analysis.md

Workflow

  1. Accept a JD URL, pasted JD text, or mixed notes.
  2. Parse deterministic facts only:
    • source URL or source type
    • accessed date when browsing or checking a live posting
    • company
    • role
    • seniority
    • deadline
    • required documents
    • cover letter questions and length limits
  3. Analyze explicit requirements and inferred evaluation criteria in separate sections. Read the JD alone first; fix mandatory, core, and preferred importance and source excerpts before reading the resume. Do not adjust importance to suit the candidate.
  4. Gap map the JD against resume.md when resume evidence is available. Link each requirement to stable evidence IDs and original locations; classify it as met, partially met, unmet, or needs confirmation.
  5. Identify the likely evidence needed from resume.md.
  6. Mark gaps where the resume evidence appears weak or missing.
  7. Keep inference conservative. Label inferred criteria as inferred.

Report employment type, location, selected sub-role, stated recruitment stages, questions, counting rules, attachments, missing source text, and contradictions. Keep eligibility, fit, and preparation effort distinct. Evaluate graduation timing, career length, certificate or language validity only when the JD demands them. An unmet mandatory qualification cannot be offset by a high fit score. Recommend starting writing, supplying information, or holding, with reasons. For multi-role postings, retain posting identity separately from the selected application role. Never use another role's questions or qualifications.

Output

Use this structure:

## JD Analysis
- Company:
- Source:
- Accessed Date:
- Role:
- Career Level:
- Main Responsibilities:
- Must-Have:
- Nice-To-Have:
- Hidden Evaluation Criteria:
- Resume Keywords:
- Cover Letter Angles:
- Risk/Gaps:
- Unknown Or Unverified Fields:
- Questions To Ask:

New Grad vs Experienced

  • For new grads, look for learning ability, project relevance, tool familiarity, collaboration, and job motivation.
  • For experienced candidates, look for role scope, measurable impact, domain depth, stakeholder work, and repeatable contribution.

Fallback

If the JD is vague, ask for the target role, company, and any available posting text. Do not create requirements that are not in the JD or reasonably inferable from the role.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.