SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsinterview-writer

interview-writer

Data & analytics

View repository

Updated 10-15 19:57

About this skill

Interview Writer

AI interview-style content creation system - through structured interviews, generate original content that matches the user's style.

Core Concept

  • Not automatic writing: AI does not generate content out of thin air
  • Not ghostwriting: It is not simply recording what the user says
  • It is interview-style creation: AI analysis + questions + user answers + AI integrating according to style

Resource Loading Policy

Priority order(user configuration overrides default):

  1. User-defined(if present):~/.claude/content-profile/
  2. Skill default(as fallback): this skill's references/ directory

Loading logic：

For each profile file (writing-style, opinions, thinking-patterns, domain-knowledge):
  1. Check if ~/.claude/content-profile/{file}.md exists
  2. If exists → use user's version
  3. If not exists → use references/{file}.md from this skill

Profile update target：

  • Always update to ~/.claude/content-profile/
  • If this directory does not exist, create it

Workflow

Phase 1: Trigger analysis

Receive content trigger(the user provides any of the following):

  • An article/link (needs commentary or response)
  • An event/topic (needs an opinion expressed)
  • A product/technology (needs introduction or analysis)
  • An idea draft (needs to be expanded into an article)

AI analysis steps：

  1. Load the user profile according to the Resource Loading Policy
  2. Read ~/.claude/profile/ relevant background files under this directory (if necessary)
  3. Analyze the core issues of the triggering content
  4. Based on the existing profile, predict the user's likely stance and angle
  5. Automatically determine the content type(do not let the user choose; AI first determines the most likely type)

Automatic content type recognition：

Triggering content characteristics Determine type Output combination
Own product/feature launch Release Blog Blog + social media + Chinese-English bilingual
Major events that need media coverage Press Release Press release + social media + Chinese-English bilingual
Commentary on others' articles/events Opinion/Commentary Mainly social media, optional long-form article
Technical implementation/architecture sharing Engineering Blog Blog + social media + English priority
Industry trends/insight analysis Insight Blog Blog + social media + Chinese-English bilingual

Key principle: articles, social media, and multiple languages are not a choice relationship, but areneeded simultaneously. Provide comprehensive results at once.

Phase 2: Interview Q&A

Question generation principles：

  • First use accumulated knowledge to make automatic judgments and avoid repeated questions
  • Only ask necessary questions that cannot be inferred
  • Use a mix of multiple-choice questions (lowering the barrier) and open-ended questions (obtaining depth)

Question types：

1. 立场确认（选择题）
   "关于 X 的观点，你更倾向于 A 还是 B？"

2. 角度挖掘（开放式）
   "这件事最让你在意的是什么？"

3. 深度追问（基于回答）
   "你提到 Y，能展开说说为什么？"

4. 风格确认（选择题）
   "这篇内容的语气：A) 犀利批判 B) 冷静分析 C) 热情推荐"

5. 故事/案例询问（根据文章类型）

   **Founder 观点类**：
   - 先总结 2-3 类最适合的故事方向（基于论点）
   - 让用户选择后再描述，防止跑题
   - 示例："这个观点可以用以下几类故事支撑：A) 创业早期的决策 B) 技术选型的教训 C) 团队协作的案例，有相关经历吗？"

   **Marketing 类**：
   - 优先问客户成功案例
   - 次之问自己的故事/案例
   - 示例："有没有客户因为这个功能获得明显收益的案例？"

   **审核机制**：
   - 用户提供故事后，审核是否符合论点
   - 符合 → 使用
   - 不符合 → clarify 或拒绝，不能硬塞
   - 没有合适故事 → 不加，不编造

Interview rounds：

  • Usually 2-4 rounds of Q&A are sufficient
  • 1-3 questions per round
  • End the interview once the core viewpoints and angle are clear

Phase 3: Content generation

Generation steps：

  1. Integrate the viewpoints collected from the interview
  2. Apply writing-style.md style characteristics
  3. Follow the thinking-patterns.md thinking framework
  4. Combine domain-knowledge.md professional insights
  5. Generate the complete output package in one go

Output package structure(automatically combined based on content type):

## 输出包

### 1. 博客/长文（如适用）
- 完整 Markdown
- 中文版 + 英文版

### 2. 社交媒体
- Twitter/X 版本（280字符内，或 thread 形式）
- 中文版 + 英文版

### 3. 新闻稿（如适用）
- 标准新闻稿格式
- 中文版 + 英文版

Key principles：

  • Do not let the user choose a format; provide the full set directly
  • Do not output in multiple rounds; provide the complete result in one go
  • The user can choose which version to use, but the options must be complete

Phase 4: Profile update

After the interview ends, automatically update the user profile：

# 更新目标目录（始终写入用户目录）
~/.claude/content-profile/writing-style.md    # 新发现的表达习惯
~/.claude/content-profile/opinions.md         # 新表达的观点立场
~/.claude/content-profile/thinking-patterns.md # 新观察到的思考模式
~/.claude/content-profile/domain-knowledge.md  # 新获取的领域见解

Update principles：

  • Append incrementally; do not overwrite history
  • Mark the update date
  • Extract commonalities; avoid recording one-off details
  • If a contradiction with the existing profile is found, flag it and ask the user

Key Principles

Ask less, infer more

  • If it can be inferred from the existing profile, do not ask again
  • Repeated questions reduce the user experience
  • Each interview should be more precise than the last

Viewpoints first

  • The core value of the content is the user's unique viewpoints
  • AI handles structure and expression; the user handles thought
  • Avoid generating content that is "correct but has no viewpoint"

Consistent style

  • The generated content should read as if the user wrote it
  • Maintain the user's expression habits and tone
  • Do not beautify or "standardize" the user's style

Continuous learning

  • Every creation is an opportunity to learn about the user
  • The profile should become increasingly precise
  • Ultimate goal: only confirmation needed, no interview required

Usage Examples

Example 1: Comment on a technical article

用户: 帮我写一篇对这篇文章的评论 [提供链接]

AI: [读取画像]
    [分析文章核心观点]

    根据你之前的观点，你可能会对以下几点有看法：
    1. 作者说 X，但这和你的 Y 理念似乎有冲突
    2. 文章忽略了 Z，这是你通常关注的

    问题：
    1. 你对这篇文章的总体态度是？
       A) 基本认同，想补充一些
       B) 部分认同，有不同看法
       C) 不太认同，想提出反驳

    2. 最想表达的一个核心观点是什么？

用户: B，我想说的是...

AI: [继续追问细节]
    [生成评论]
    [更新画像]

Example 2: Write a product launch blog

用户: 帮我写产品 X 发布的博客

AI: [读取画像 + products/x.md]

    我了解你对 X 的定位。为了写好这篇博客：

    1. 这次发布想强调哪个方向？
       A) 技术架构创新
       B) 解决的实际问题
       C) 与其他方案的差异
       D) 愿景和长期方向

    2. 目标读者是谁？
       A) 技术开发者
       B) 产品决策者
       C) 投资者/观察者
       D) 混合

用户: A + B，主要给技术开发者看

AI: [追问技术细节重点]
    [生成博客]
    [更新画像]

Resources

Skill Default References

This skill includes default configurations (references/ directory):

  • writing-style.md - Default writing style
  • opinions.md - Default viewpoint template
  • thinking-patterns.md - Default thinking mode
  • domain-knowledge.md - Default domain knowledge

User Override Location

Users can create custom configurations in the following locations (with higher priority than defaults):

  • ~/.claude/content-profile/writing-style.md
  • ~/.claude/content-profile/opinions.md
  • ~/.claude/content-profile/thinking-patterns.md
  • ~/.claude/content-profile/domain-knowledge.md

Additional User Context

Load when needed:

  • ~/.claude/profile/ - User basic profile
  • ~/.claude/products/ - Product knowledge base