Personnel Evaluation by Humanity Test

SkillDev tools

Lets your agent assess whether a candidate can be trusted using Guan Zhong's tests based on violations of natural family bonds.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Personnel Evaluation by Humanity Test skill

About this capability

Use when vetting subordinates or appointees for trustworthiness. Applies Guan Zhong's humanity test (非人情) to identify dangerous individuals who violate natural bonds — parental, family, and self-preservation — to gain power.

What this skill tells your AI

The instructions your AI receives, as published by baojie/shiji-kb in kg/ontology/ontology-v1/skus/procedural/skill_071/SKILL.md and read by ahel’s review.

Overview

A method attributed to Guan Zhong for identifying untrustworthy individuals by examining whether their actions violate basic human nature and natural affections.

Core Principle

People who violate fundamental human bonds to please superiors cannot be trusted. Their extreme actions indicate they seek personal gain at any cost.

Evaluation Criteria

Test 1: Parental Bond Test

  • Question: Has the person harmed their own children to please a superior?
  • Example: Yi Ya (易牙) killed his own son to cook for the ruler
  • Verdict: "非人情,不可" (Against human nature, cannot be trusted)

Test 2: Family Loyalty Test

  • Question: Has the person abandoned or betrayed their parents/close family for a superior?
  • Example: Kai Fang (开方) left his parents to serve
  • Verdict: "非人情,难近" (Against human nature, difficult to approach safely)

Test 3: Self-Preservation Test

  • Question: Has the person mutilated themselves to please a superior?
  • Example: Shu Diao (竖刀) castrated himself to serve in the palace
  • Verdict: "非人情,难亲" (Against human nature, difficult to trust closely)

Decision Matrix

Violation TypeTrust LevelRecommendation
Harmed childNoneNever appoint
Abandoned familyVery LowKeep at distance
Self-mutilationLowDo not give power
None of aboveEvaluate furtherProceed with other tests

Expected Outcomes

  • Identification of potentially treacherous individuals before they gain power
  • Protection of the state from those who lack natural human bonds
  • More reliable leadership team

Validation

  1. Confirm each candidate is evaluated against all three tests: parental bond (Yi Ya model), family loyalty (Kai Fang model), and self-preservation (Shu Diao model)
  2. Verify that anyone flagged as 非人情 is excluded from positions of power regardless of their apparent usefulness
  3. Check that candidates who pass all three tests proceed to further evaluation rather than being automatically trusted

Signals

GitHub stars
3k
Forks
425
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
personnel-evaluation-by-humanity
Source
github.com/baojie/shiji-kb