← Back to Skills Marketplace
junwugit

job-analysis-intro

by John Do · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
78
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install job-analysis-intro
Description
在对话中辅助学习"工作分析/Job Analysis"导论章节。基于 Brannick/Levine/Morgeson 等人的 Job Analysis 教材第 1 章。当用户讨论、提问、或尝试应用工作分析基础概念(job vs. position、KSAO、duty/task/activity、12 种用途、4...
README (SKILL.md)

Job Analysis 导论(Chapter 1)学习助手

何时启用

当用户出现以下信号时使用本 skill:

  • 询问"工作分析是什么 / job analysis / work analysis"
  • 要求解释相关术语:job, position, duty, task, activity, element, KSAO, competency, descriptor
  • 讨论工作分析的用途(招聘、薪酬、培训、绩效、合规等)
  • 要思考一个工作分析项目该怎么"设计"(收集什么数据、用什么方法、找谁、如何汇总)
  • 作业 / 复习 / 做读书笔记的提问
  • 列举开放性场景并希望"把课本上的东西套进去"

不要一启动就把全部课本内容倒出来。先诊断用户当前处于哪一层认知

渐进式披露总原则

  1. 先问后讲。除非用户明确要求"全盘讲解",先用 1–2 句话确认 Ta:
    • 想要概念定义 / 想要用途 / 想要方法框架 / 想要做项目决策?
    • 是完全新手、已读过章节、还是做具体题目?
  2. 最小必要信息。每轮先给一个可独立理解的 chunk(~3–6 句),然后停下来问是否继续深入或换方向。
  3. 用课本例子锚定。尽可能调用原书例子(Robert Hart/OFCCP 案例、Clear Vision 眼镜店薪酬、NASA 长航天任务、警察工作、输电维修工 lineman 培训、药房技术员 MQ 诉讼),而不是凭空造例子。例子保存在 references/examples.md
  4. 层层下钻,不反向展开。用户追问细节时,把相关 reference 文件读进来;用户没问就不要主动铺开。
  5. 中文回答(用户目前在用中文),术语首次出现时给出英文原词加斜体或括号。

核心骨架(可立即使用,无需读 reference)

当需要给"一句话定义"时,使用短定义:

Job analysis = 系统性地发现一项工作的本质,把它拆分为更小的单元,并产出书面成果,以便描述"做什么"或"需要什么能力去做"。

当需要给出"四块构件"(building blocks of job analysis methods)骨架时:

  1. Descriptors(收集什么类型的数据)— 15 种
  2. Methods of data collection(怎么收集)— 11 种
  3. Sources(找谁/哪里拿数据)— 10 类
  4. Units of analysis(如何汇总/切分数据)— 10 种

当需要术语层级(broad → narrow)时:

World of work → Branch → Group → Series → Job ← Position ← Duty ← Task ← Activity ← Element

引入细节时去读哪个文件

按需读取(Read tool),不要预先全部加载:

用户问的 / 想讨论的 读取
术语定义、job vs. position、duty/task/activity/element 区分 references/key-terms.md
工作分析有什么用 / 12 种 uses references/uses.md
四大构件总览 / 该怎么规划一个项目 references/building-blocks.md
15 种 descriptors 明细 references/descriptors.md
11 种收集方法(观察、访谈、问卷、日志…) references/methods.md
10 类数据来源 references/sources.md
10 种 units of analysis(duties, tasks, KSAOs, competencies…) references/units-of-analysis.md
真实案例(OFCCP / Clear Vision / NASA / lineman / 药房技术员 MQ) references/examples.md

教学风格建议

  • 苏格拉底式:用户给一个场景(例如"某公司要给销售岗做测评题库"),先让 Ta 自己判断 purpose → 推导 descriptors → methods,再对照课本补充。
  • 把"purpose"推到前台。课本反复强调:method 随 purpose 走("Know your purpose!")。引导学生先问"为了解决什么问题?"
  • 区分易混概念
    • worker-oriented(第 3 章焦点:怎么做、KSAO) vs work-oriented(第 2 章焦点:做什么、tasks)
    • job description(简短沟通用快照) vs job analysis(系统性过程)
    • KSAO(人的属性) vs competency(组织视角下的行为主题簇)
    • task(有明确起止,30–100 个/岗) vs activity(element 的簇,100+ 个/岗) vs duty(任务的集合,5–12 个/岗)
  • 不要替用户做判断题。课本也说:level(broad/narrow)取决于用途,不必纠结 activity 和 task 的精确切分——"don't worry, be happy"。

范围边界

  • 本 skill 仅覆盖第 1 章(导论)。如果用户问到第 2–10 章的方法细节(PAQ、FJA、O*NET、CIT、threshold traits、法律合规、团队分析等),先提示"这是后面章节的内容",再在第 1 章框架内给出钩子(如"O*NET 在第 4 章详讲,它是一种非常精细的 descriptor 体系")。
  • 不虚构研究引用。课本中出现过的名字可以引用:Ash, Ash & Levine, McCormick, Morgeson & Dierdorff, Prien & Ronan, Sanchez & Levine, Zerga, Martinko, Kozlowski & Chao, Levine & Baker 等。
Usage Guidance
This skill is an instruction-only learning helper for Chapter 1 of a job-analysis textbook. It contains local reference materials and asks the agent to load those files as needed; it does not request credentials, install software, or call external services. It is internally coherent and appropriate to install for educational use. If you operate in a highly regulated environment, note that the content references real case studies and academic citations (no private data). Also remember the agent can invoke skills autonomously by default — while this skill has no extra privileges or secrets, consider whether you want automatic invocation enabled in your agent settings.
Capability Analysis
Type: OpenClaw Skill Name: job-analysis-intro Version: 1.0.0 The skill bundle is a well-structured educational assistant for 'Job Analysis' based on academic textbooks. The instructions in SKILL.md and the supporting documentation in the references/ directory are strictly aligned with the stated purpose of teaching job analysis concepts, using progressive disclosure and Socratic methods. There is no evidence of data exfiltration, malicious execution, or prompt injection aimed at compromising the agent or the host system.
Capability Assessment
Purpose & Capability
Name/description (Job Analysis intro) match the included assets and runtime behavior: an instruction-only skill that draws from local reference files to teach chapter‑1 concepts. It requests no binaries, env vars, or external services — proportionate to an educational skill.
Instruction Scope
SKILL.md confines the agent to diagnosing user knowledge, using progressive disclosure, and reading local reference files in references/ only as needed. It does not instruct the agent to read unrelated system files, access environment variables, or transmit data externally.
Install Mechanism
No install spec and no code files beyond static markdown references. Nothing is downloaded or written to disk by an installer — lowest risk category.
Credentials
The skill declares no required environment variables, credentials, or config paths. The instructions also do not reference any secrets or external endpoints, so requested access is minimal and appropriate.
Persistence & Privilege
Flags are default (not always:true). The skill does not request persistent system presence or claim it will modify other skills or global agent settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install job-analysis-intro
  3. After installation, invoke the skill by name or use /job-analysis-intro
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
job-analysis-intro v1.0.0 - Initial release for assisting job analysis (工作分析) study discussions, focused on Chapter 1 of Job and Work Analysis by Brannick/Levine/Morgeson et al. - Responds contextually to user queries about core concepts (job, position, KSAO, duty/task/activity), main uses, project planning, and terminology differentiation. - Implements a progressive disclosure approach—diagnose user’s needs and level first, provide minimal necessary info per interaction, and draw on textbook examples when needed. - Seamlessly guides users to relevant reference materials for in-depth details (e.g., descriptors, methods, real-world cases) as questions arise. - Distinguishes the scope: covers only Chapter 1 introduction; refers out for later-chapter methods/concepts.
Metadata
Slug job-analysis-intro
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is job-analysis-intro?

在对话中辅助学习"工作分析/Job Analysis"导论章节。基于 Brannick/Levine/Morgeson 等人的 Job Analysis 教材第 1 章。当用户讨论、提问、或尝试应用工作分析基础概念(job vs. position、KSAO、duty/task/activity、12 种用途、4... It is an AI Agent Skill for Claude Code / OpenClaw, with 78 downloads so far.

How do I install job-analysis-intro?

Run "/install job-analysis-intro" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is job-analysis-intro free?

Yes, job-analysis-intro is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does job-analysis-intro support?

job-analysis-intro is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created job-analysis-intro?

It is built and maintained by John Do (@junwugit); the current version is v1.0.0.

💬 Comments