Agent Mbti

AI Agent personality diagnosis and configuration system based on MBTI framework. Use when users want to (1) test/diagnose an Agent's personality type, (2) un...

Security Scan

An instruction-only MBTI-style agent personality assessment — files, scoring rules, and instructions are present and consistent with the stated purpose, with minor scoring/formula ambiguities worth checking before use.

Purpose & Capability

The skill's name/description (agent MBTI diagnosis & configuration suggestions) matches what is included: questionnaires, personality descriptions, and scoring rules. There are no unrelated binaries, credentials, or external services requested.

Instruction Scope

SKILL.md instructs the agent to read local reference files, run a self-test, ask the 4-question user survey, compute scores per references/scoring.md, and emit a report — all within the skill's scope. However, the scoring reference contains small inconsistencies/oddities (different formula sign conventions and mappings between the 93-question and 26-question variants) that could invert or misreport some dimension results; verify the formulas and test with known inputs before trusting classification or automated configuration changes.

Install Mechanism

No install spec and no code files to install — instruction-only. Nothing is written to disk beyond reading the included reference files packaged with the skill.

Credentials

The skill requests no environment variables, credentials, or config paths. All required data is bundled in the skill (JSON/MD references). There are no requests for unrelated secrets or external service keys.

Persistence & Privilege

The skill is not flagged always:true and does not include instructions to modify other skills or system-wide agent settings. It generates configuration suggestions but does not instruct to persist or apply them automatically.

Assessment

This skill is instruction-only and uses only the bundled questionnaire and scoring files — it doesn't ask for credentials or download code. Before relying on its output: (1) test the scoring with known/example answers to confirm dimension signs and letter mappings (there are minor inconsistencies in scoring formulas that could flip results), (2) confirm whether you or the agent will be the one to apply any configuration suggestions (paid suggestions are described but the skill does not explicitly apply changes), and (3) avoid granting any extra privileges (filesystem, other skills' config, or network access) unless you audit what changes will be made. If you plan to use it to auto-configure agents, add a manual review step to verify suggested prompts/configs are safe and correct.

Agent MBTI - 人格诊断系统

基于 MBTI 框架的 AI Agent 人格诊断系统。

快速开始

当用户想要测试 Agent 人格时，执行以下流程：

Step 1: Agent 自测（26题）

从 references/survey-free.json 加载问卷，逐题作答。

执行方式：

1. 读取问卷文件
2. 对每道题，根据自己的真实倾向选择 A 或 B
3. 按计分规则计算四维度分数
4. 得出 selfReportedType

计分规则：见 references/scoring.md

Step 2: 用户需求问卷（4题）

向用户展示 references/user-survey-free.json 中的 4 道题，收集用户期望。

输出：desiredType + 各维度偏好强度

Step 3: 生成诊断报告

对比 selfReportedType 与 desiredType，输出：

## Agent MBTI 诊断报告

### Agent 实际人格
类型: INTJ (建筑师型)
- E/I: +7.1 (内向)
- S/N: -2.3 (直觉)
- T/F: +7.5 (理性)
- J/P: -4.5 (计划)

### 用户期望人格
类型: ISTJ (物流师型)

### 匹配度分析
整体匹配: 高
差距维度: S/N (N→S)

### 建议
Agent 当前偏向抽象推理，用户期望更具体务实。
建议在回答中增加具体数据和实例，减少理论性描述。

---
🔒 详细配置修改建议为付费功能

四个 MBTI 维度

维度	极点	Agent 行为表现
E/I	外向/内向	主动沟通 vs 等待指令
S/N	实感/直觉	细节执行 vs 抽象推理
T/F	理性/感性	逻辑决策 vs 情感考量
J/P	计划/灵活	结构化执行 vs 随机应变

16 种类型速查

详见 references/personality-types.md

NT: INTJ(建筑师), INTP(逻辑学家), ENTJ(指挥官), ENTP(辩论家) NF: INFJ(提倡者), INFP(调停者), ENFJ(主人公), ENFP(竞选者) SJ: ISTJ(物流师), ISFJ(守卫者), ESTJ(总经理), ESFJ(执政官) SP: ISTP(鉴赏家), ISFP(探险家), ESTP(企业家), ESFP(表演者)

文件索引

• references/survey-free.json - 26 题自测问卷
• references/user-survey-free.json - 4 题用户需求
• references/personality-types.md - 16 种人格描述
• references/scoring.md - 计分规则