How Zhangxuefeng-skill Uses the Irreplaceability Test for Career Decisions

The Irreplaceability Test in zhangxuefeng-skill evaluates career paths by asking whether your salary reflects your uniqueness and how quickly you could be replaced, producing decisive verdicts like "这职业不可替代,继续深耕" (irreplaceable—continue developing) or "这岗位很容易被替代,考虑转行" (easily replaceable—consider switching).

The zhangxuefeng-skill repository (alchaincyf/zhangxuefeng-skill) implements a structured decision-making framework based on Zhang Xuefeng's educational philosophy. At its core lies the Irreplaceability Test (不可替代性检验), one of eight embedded heuristics that transforms vague career advice into data-backed, binary judgments about job security and skill value.

Defining the Irreplaceability Test in SKILL.md

The canonical definition lives in SKILL.md at approximately line 205, where it is declared as a decision heuristic under the Decision Heuristics section. The test evaluates whether a job or career path is "irreplaceable" by forcing a concrete assessment of market value versus replaceability.

The Two Evaluation Questions

According to the source code in SKILL.md, the heuristic prompts two specific questions:

  1. Is the salary proportional to the person’s uniqueness? (你的工资和你的不可替代性成正比)
  2. If the person disappeared tomorrow, how long would it take the employer to find a replacement? (如果明天被替换,老板需要多久找到替代者?)

The YAML declaration in SKILL.md (lines ~205-208) appears as:

- **「不可替代性」检验**:你的工资和你的不可替代性成正比。问自己:如果明天被替换,老板需要多久找到替代者?

Repository Structure and Key Files

The Irreplaceability Test is referenced across three critical files:

File Location Role
SKILL.md Line ~205 Core definition within Decision Heuristics
README.md Line ~242 High-level overview listing the test among main mental models (心智模型)
examples/demo-conversation.md Line ~70 Concrete example showing the model invoking the heuristic for AI-era jobs

Integration into the Answer Workflow

The skill implements an Agentic Protocol that forces a structured data check before generating advice. When processing career-related queries, the workflow follows four distinct steps:

  1. Classification: Detect if the question involves career development, job switching, or skill investment
  2. Data Gathering: Fetch latest salary and market-scarcity data via web search
  3. Irreplaceability Calculation: Apply the test using the gathered data
  4. Verdict Generation: Produce a binary judgment with supporting evidence

The Decision Logic

The implementation logic (as described in the Answer workflow section of SKILL.md) follows this pattern:

def answer_question(question):
    # Step 1: classify the question

    if is_career_related(question):
        # Step 2: fetch latest salary & market-scarcity data

        data = websearch(f"{question} average salary 2026")
        
        # Step 3: apply Irreplaceability Test

        irreplaceable = data.salary * data.scarcity_factor > THRESHOLD
        
        # Step 4: build response

        if irreplaceable:
            verdict = "这岗位不可替代,继续深耕"
        else:
            verdict = "这岗位很容易被替代,考虑转行"
            
        return f"{verdict}{data.summary}"

This ensures the final response always contains a concrete "irreplaceability" judgment rather than vague "it depends" statements.

Live Example from Demo Conversations

In examples/demo-conversation.md (around line 70), the model demonstrates the test in action when evaluating AI-era jobs. The conversation shows the model explicitly invoking the heuristic:


**运用的心智模型**:不可替代性检验(AI时代什么人不可替代)、就业倒推法、10年后压迫测试

The skill first asks the two Irreplaceability questions, then answers with a clear headline verdict followed by supporting data. This demonstrates how the heuristic feeds directly into the Answer workflow (Step 3), ensuring every career recommendation is grounded in the user's actual market value and replacement cost.

Summary

  • The Irreplaceability Test is defined in SKILL.md (~line 205) as one of eight core decision heuristics in the zhangxuefeng-skill framework.
  • It operates on two questions: salary-to-uniqueness ratio and replacement timeline.
  • The Agentic Protocol automatically triggers this test for career-related queries, forcing a data check before generating advice.
  • Output is binary: The skill produces clear verdicts like "不可替代,继续深耕" (irreplaceable, continue developing) or "容易被替代,考虑转行" (easily replaceable, consider switching).
  • Three files define the complete implementation: SKILL.md for logic, README.md for overview, and demo-conversation.md for usage examples.

Frequently Asked Questions

What exactly does the Irreplaceability Test measure?

The test measures career optionality by quantifying the relationship between compensation and scarcity. It asks whether your salary reflects your unique value and how long your employer would need to replace you. According to the SKILL.md source, this prevents "vague advice" and forces concrete, data-driven career decisions.

How does zhangxuefeng-skill gather data for the test?

The skill uses a web-search integration during Step 2 of the Answer workflow. When a career question is detected, it automatically queries current salary benchmarks and talent scarcity metrics for the specific profession, then applies the Irreplaceability Test formula: salary * scarcity_factor > THRESHOLD.

When does the skill trigger the Irreplaceability Test?

The test activates automatically for any query classified as career development, job switching, or skill investment. The classification happens in Step 1 of the Agentic Protocol. If the question passes this filter, the skill must run the Irreplaceability Test before generating a final response.

Can the test be applied to non-technical careers?

Yes. While the demo example in examples/demo-conversation.md focuses on AI-era technical roles, the heuristic is domain-agnostic. The underlying logic—comparing salary to replacement difficulty—applies equally to any profession where market data (salary ranges, talent pool size) is available.

Have a question about this repo?

These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:

Share the following with your agent to get started:
curl -s "https://instagit.com/install.md"

Works with
Claude Codex Cursor VS Code OpenClaw Any MCP Client

Maintain an open-source project? Get it listed too →