personality-profiler

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Personality Profiler

人格档案生成器

Generate comprehensive, extensible personality profiles from social media data exports.

从社交媒体数据导出文件生成全面、可扩展的人格档案。

Overview

概述

This skill analyzes exported social media data to create detailed personality profiles suitable for:

AI assistant personalization (training data for personalized responses)
Self-reflection and pattern discovery

本技能分析导出的社交媒体数据，创建详细的人格档案，适用于：

AI助手个性化（用于生成个性化回复的训练数据）
自我反思与模式发现

Workflow

工作流程

Receive data — User provides exported data files (JSON/CSV)
Parse data — Extract posts, comments, interactions using platform-specific parsers
Analyze dimensions — Evaluate across 8 personality dimensions
Generate profile — Output structured profile in extensible JSON format
Summarize insights — Provide human-readable summary

接收数据 — 用户提供导出的数据文件（JSON/CSV格式）
解析数据 — 使用平台专属解析器提取帖子、评论和互动内容
维度分析 — 从8个人格维度进行评估
生成档案 — 以可扩展的JSON格式输出结构化档案
洞察总结 — 提供易读的人工总结

Supported Platforms

支持的平台

Platform	Export Type	Key Files
Twitter/X	ZIP archive	`tweets.js` , `like.js` , `profile.js`
LinkedIn	ZIP archive	`Profile.csv` , `Connections.csv` , `Comments.csv` , `Shares.csv`
Instagram	ZIP archive	`content/posts_1.json` , `comments.json` , `profile.json`

For detailed format specifications, see references/platform-formats.md.

平台	导出类型	关键文件
Twitter/X	ZIP压缩包	`tweets.js` , `like.js` , `profile.js`
LinkedIn	ZIP压缩包	`Profile.csv` , `Connections.csv` , `Comments.csv` , `Shares.csv`
Instagram	ZIP压缩包	`content/posts_1.json` , `comments.json` , `profile.json`

如需详细格式规范，请查看references/platform-formats.md。

Analysis Dimensions

分析维度

Analyze content across these 8 dimensions:

从以下8个维度分析内容：

1. Communication Style

1. 沟通风格

Tone: formal ↔ casual, serious ↔ playful, direct ↔ diplomatic
Verbosity: concise ↔ elaborate, uses bullet points vs paragraphs
Vocabulary: technical level, industry jargon, colloquialisms

语气: 正式 ↔ 随意，严肃 ↔ 活泼，直接 ↔ 委婉
话语量: 简洁 ↔ 详尽，使用项目符号 vs 段落
词汇: 技术水平、行业术语、口语表达

2. Interests & Expertise

2. 兴趣与专长

Topics: recurring themes, domains of focus
Depth: surface mentions vs deep engagement
Evolution: how interests have changed over time

主题: 重复出现的话题、关注领域
深度: 表面提及 vs 深度参与
演变: 兴趣随时间的变化

3. Values & Beliefs

3. 价值观与信念

Priorities: what matters most (inferred from emphasis)
Advocacy: causes supported or promoted
Philosophy: worldview indicators

优先级: 最关注的事物（从内容侧重点推断）
倡导: 支持或推广的事业
理念: 世界观相关迹象

4. Social Patterns

4. 社交模式

Engagement style: initiator vs responder, commenter vs creator
Network orientation: broad reach vs tight community
Interaction tone: supportive, challenging, neutral

参与风格: 发起者 vs 回应者，评论者 vs 创作者
网络导向: 广泛触达 vs 紧密社群
互动语气: 支持性、挑战性、中立

5. Emotional Expression

5. 情绪表达

Range: emotional vocabulary breadth
Valence: positive/negative tendency
Triggers: what elicits strong reactions

范围: 情绪词汇的广度
倾向: 积极/消极倾向
触发点: 引发强烈反应的事物

6. Cognitive Style

6. 认知风格

Reasoning: analytical vs intuitive, data-driven vs narrative
Complexity: nuanced vs straightforward positions
Openness: receptivity to new ideas

推理方式: 分析型 vs 直觉型，数据驱动 vs 叙事型
复杂度: 立场的细致度 vs 直白度
开放性: 对新想法的接受度

7. Professional Identity

7. 职业身份

Domain: industry, role, expertise areas
Aspirations: career direction signals
Network: professional relationship patterns

领域: 行业、职位、专长领域
抱负: 职业发展方向的信号
人脉: 职业关系模式

8. Temporal Patterns

8. 时间模式

Activity rhythms: when they post, reply, engage
Content cycles: seasonal or event-driven patterns
Growth trajectory: how expression has evolved

活动节奏: 发帖、回复、互动的时间规律
内容周期: 季节性或事件驱动的模式
成长轨迹: 表达方式的演变

Profile Schema

档案Schema

Output profiles in this extensible JSON structure:

json

{
  "version": "1.0",
  "generated_at": "ISO-8601 timestamp",
  "data_sources": [
    {
      "platform": "twitter|linkedin|instagram",
      "date_range": {"start": "YYYY-MM-DD", "end": "YYYY-MM-DD"},
      "item_count": 1234
    }
  ],
  "profile": {
    "summary": "2-3 paragraph narrative summary",
    "dimensions": {
      "communication_style": {
        "confidence": 0.0-1.0,
        "traits": {
          "formality": {"value": -1.0 to 1.0, "evidence": ["quote1", "quote2"]},
          "verbosity": {"value": -1.0 to 1.0, "evidence": []},
          "directness": {"value": -1.0 to 1.0, "evidence": []}
        },
        "patterns": ["pattern1", "pattern2"],
        "recommendations_for_ai": "How an AI should communicate with this person"
      }
    },
    "notable_quotes": [
      {"text": "quote", "context": "why notable", "dimension": "which dimension"}
    ],
    "keywords": ["term1", "term2"],
    "topics_ranked": [
      {"topic": "name", "frequency": 0.0-1.0, "sentiment": -1.0 to 1.0}
    ]
  },
  "extensions": {}
}

The

extensions

field allows adding custom dimensions without breaking compatibility.

以可扩展的JSON结构输出档案：

json

{
  "version": "1.0",
  "generated_at": "ISO-8601 timestamp",
  "data_sources": [
    {
      "platform": "twitter|linkedin|instagram",
      "date_range": {"start": "YYYY-MM-DD", "end": "YYYY-MM-DD"},
      "item_count": 1234
    }
  ],
  "profile": {
    "summary": "2-3 paragraph narrative summary",
    "dimensions": {
      "communication_style": {
        "confidence": 0.0-1.0,
        "traits": {
          "formality": {"value": -1.0 to 1.0, "evidence": ["quote1", "quote2"]},
          "verbosity": {"value": -1.0 to 1.0, "evidence": []},
          "directness": {"value": -1.0 to 1.0, "evidence": []}
        },
        "patterns": ["pattern1", "pattern2"],
        "recommendations_for_ai": "How an AI should communicate with this person"
      }
    },
    "notable_quotes": [
      {"text": "quote", "context": "why notable", "dimension": "which dimension"}
    ],
    "keywords": ["term1", "term2"],
    "topics_ranked": [
      {"topic": "name", "frequency": 0.0-1.0, "sentiment": -1.0 to 1.0}
    ]
  },
  "extensions": {}
}

extensions

字段允许添加自定义维度，且不会破坏兼容性。

Process

处理流程

Step 1: Data Ingestion

步骤1：数据导入

When user provides files:

Identify platform from file structure
Locate key content files (see platform table above)
Parse using appropriate format handler
Normalize to common internal structure:

json

{
  "items": [
    {
      "id": "unique_id",
      "type": "post|comment|share|like",
      "timestamp": "ISO-8601",
      "content": "text content",
      "metadata": {
        "platform": "twitter",
        "engagement": {"likes": 0, "replies": 0, "shares": 0},
        "context": "reply_to_id or null"
      }
    }
  ]
}

当用户提供文件时：

根据文件结构识别平台
定位关键内容文件（见上方平台表格）
使用对应的格式处理器解析
标准化为通用内部结构：

json

{
  "items": [
    {
      "id": "unique_id",
      "type": "post|comment|share|like",
      "timestamp": "ISO-8601",
      "content": "text content",
      "metadata": {
        "platform": "twitter",
        "engagement": {"likes": 0, "replies": 0, "shares": 0},
        "context": "reply_to_id or null"
      }
    }
  ]
}

Step 2: Content Analysis

步骤2：内容分析

For each dimension:

Extract signals — Find relevant content snippets
Score traits — Rate on dimension-specific scales
Gather evidence — Collect representative quotes
Calculate confidence — Based on data volume and consistency

Minimum thresholds for confident analysis:

50+ posts for basic profile
200+ posts for detailed profile
500+ posts for high-confidence profile

If below thresholds, note reduced confidence in output.

针对每个维度：

提取信号 — 找到相关内容片段
特质评分 — 按维度专属量表打分
收集证据 — 收集代表性引用
计算置信度 — 根据数据量和一致性计算

可信分析的最低阈值：

50篇以上帖子：基础档案
200篇以上帖子：详细档案
500篇以上帖子：高置信度档案

若低于阈值，需在输出中注明置信度降低。

Step 3: Profile Generation

步骤3：档案生成

Populate all dimension objects in schema
Write narrative summary synthesizing key findings
Extract notable quotes (5-10 most characteristic)
Rank topics by frequency and engagement
Generate AI personalization recommendations

填充Schema中的所有维度对象
撰写综合关键发现的叙事性总结
提取代表性引用（5-10条最具特征的内容）
按频率和互动量排序主题
生成AI个性化建议

Step 4: Output Delivery

步骤4：输出交付

Provide two outputs:

JSON profile — Complete structured data (save as
```
personality_profile.json
```
)
Markdown summary — Human-readable insights document

提供两种输出：

JSON档案 — 完整的结构化数据（保存为
```
personality_profile.json
```
）
Markdown总结 — 易读的洞察文档

AI Personalization Recommendations

AI个性化建议

For each dimension, include specific guidance for AI systems:

Example recommendations:

communication_style.recommendations_for_ai:
"Use a conversational but informed tone. Avoid excessive formality.
Include occasional humor. Lead with conclusions, then supporting detail.
Match their tendency for medium-length responses (2-3 paragraphs)."

interests.recommendations_for_ai:
"Can reference machine learning, distributed systems, and startup culture
without explanation. Assume familiarity with Python ecosystem. May enjoy
tangential connections to philosophy of technology."

针对每个维度，为AI系统提供具体指导：

示例建议：

communication_style.recommendations_for_ai:
"Use a conversational but informed tone. Avoid excessive formality.
Include occasional humor. Lead with conclusions, then supporting detail.
Match their tendency for medium-length responses (2-3 paragraphs)."

interests.recommendations_for_ai:
"Can reference machine learning, distributed systems, and startup culture
without explanation. Assume familiarity with Python ecosystem. May enjoy
tangential connections to philosophy of technology."

Handling Multiple Platforms

多平台处理

When analyzing data from multiple platforms:

Process each platform separately first
Cross-reference for consistency
Note platform-specific behaviors (e.g., more formal on LinkedIn)
Weight professional platforms for work identity
Weight personal platforms for authentic voice
Merge into unified profile with platform annotations

当分析来自多个平台的数据时：

先分别处理每个平台的数据
交叉验证一致性
标注平台专属行为（如LinkedIn上更正式）
加权职业平台数据以体现职业身份
加权个人平台数据以体现真实表达
合并为带平台标注的统一档案

Privacy Considerations

隐私注意事项

Before processing:

Confirm user owns the data
Note that analysis stays local (no external API calls for content)
Offer to redact specific people/topics if requested
Output can be edited before use

处理前：

确认用户拥有数据所有权
说明分析仅在本地进行（不会调用外部API处理内容）
若用户要求，可提供特定人物/话题的脱敏功能
输出内容可在使用前编辑

Extending the Profile

扩展档案

The profile schema supports extensions:

json

{
  "extensions": {
    "custom_dimension": {
      "confidence": 0.8,
      "traits": {},
      "patterns": [],
      "recommendations_for_ai": ""
    },
    "domain_specific": {
      "developer_profile": {
        "languages": ["python", "rust"],
        "paradigm_preference": "functional-leaning"
      }
    }
  }
}

Users can request custom dimensions by describing what they want analyzed.

档案Schema支持扩展：

json

{
  "extensions": {
    "custom_dimension": {
      "confidence": 0.8,
      "traits": {},
      "patterns": [],
      "recommendations_for_ai": ""
    },
    "domain_specific": {
      "developer_profile": {
        "languages": ["python", "rust"],
        "paradigm_preference": "functional-leaning"
      }
    }
  }
}

用户可通过描述需求，请求添加自定义分析维度。