writer

What this skill does

# Writer Skill

Generate content in your authentic voice by analyzing writing samples and applying learned voice patterns through adaptive interviewing and voice-matched generation.

## Philosophy

**Key Innovation**: This isn't a generic AI writer - it's YOUR voice, learned from YOUR writing.

Unlike generic AI content generation, this skill:
- Learns from 6 distinct voice profiles (work email, personal email, blog, LinkedIn, Twitter, reports)
- Interviews you like a Pulitzer-winning journalist to deeply understand your message
- Validates output against your authentic patterns (not generic "good writing")
- Saves versioned drafts to Obsidian for iteration and history

**Simplicity**: Geoffrey's native LLM capabilities analyze your voice. Scripts just fetch samples. No complex NLP libraries - clean, maintainable architecture.

---

## When to Activate

Use this skill when you need to:
- Write emails that sound like you (not generic AI)
- Draft blog posts in your established voice
- Create social media content consistent with your style
- Generate reports matching your professional tone
- Ensure content authentically represents your perspective and voice

**DON'T use for**:
- Quick factual responses (use native Claude)
- Content in someone else's voice
- Generic templates

---

## Voice Profiles

Six distinct profiles capture different contexts:

| Profile | Source | Sample Target | Status |
|---------|--------|---------------|--------|
| `email_work` | PSD Gmail (sent emails, 6mo) | 50+ emails | Check writing-voice.json |
| `email_personal` | Personal/HRG Gmail (sent, 6mo) | 50+ emails | Check writing-voice.json |
| `blog_technical` | psd401.ai + blog.krishagel.com | 10+ posts | Check writing-voice.json |
| `social_linkedin` | LinkedIn profile posts | 30+ posts | Check writing-voice.json |
| `social_twitter` | X/Twitter posts | 30+ posts | Check writing-voice.json |
| `report_formal` | User-provided samples | 5-10 reports | Check writing-voice.json |

**Confidence Levels**:
- **High (0.85+)**: 50+ samples - voice well established
- **Moderate (0.70-0.84)**: 20-49 samples - patterns emerging
- **Low (<0.70)**: <20 samples - insufficient data, warn user

---

## 6-Phase Workflow

### Phase 1: Voice Profile Selection & Validation

**Goal**: Load the right voice profile and verify it's ready

**Process**:
1. Detect content type from user request:
   - "write email" → `email_work` or `email_personal` (ask which)
   - "blog post" → `blog_technical`
   - "LinkedIn post" → `social_linkedin`
   - "tweet" → `social_twitter`
   - "report" → `report_formal`

2. Load voice profile:
   ```bash
   cat ~/Library/Mobile\ Documents/com~apple~CloudDocs/Geoffrey/knowledge/writing-voice.json
   ```

3. Check confidence and sample count:
   - If confidence ≥ 0.85: Proceed with confidence
   - If confidence 0.70-0.84: Proceed with note
   - If confidence < 0.70: Warn user, offer options

**Low Confidence Response**:
```
Only {sample_count} {content_type} samples (confidence: {confidence}).
Target: {target_samples}+ samples for 0.85+ confidence.

Options:
1. Gather more samples (recommended)
2. Use {alternative_profile} voice as proxy (confidence: {alt_confidence})
3. Proceed anyway (may not fully match your voice)

Preference?
```

**Checkpoint 1**: User confirms profile or gathers more samples

---

### Phase 2: Deep Interview (Adaptive)

**Goal**: Deeply understand what you want to communicate

**Process**:
1. Load interview questions from `templates/interview-questions.json`

2. Adapt depth to content type:
   - **Tweet/Social**: 3-5 questions (~2-3 min)
   - **Email Quick**: 4 questions (~3 min)
   - **Email Complex**: 8 questions (~5 min)
   - **Blog**: 8-12 questions (~5-8 min)
   - **Report**: 15+ questions (~10-15 min)

3. Interview like a journalist:
   - Ask follow-up questions for vague answers
   - Probe for examples, data, evidence
   - Clarify ambiguities
   - Build structured understanding

**Email Interview Example** (complex):
```
1. Who are the recipient(s) and what's the context?
   → [User answers]

2. What's the primary goal?
   → [User answers]
   [If vague: "Can you be more specific about the outcome you want?"]

3. What background do they need?
   → [User answers]

4. What are the key points? (Max 3-5)
   → [User answers]
   [If >5: "Let's focus on the most critical 3-5 points."]

... continue through 8 questions
```

**Blog Interview Example**:
```
1. What's your central message?
2. Why does this matter right now?
3. What should readers do/think/feel after?
4. What evidence supports this?
5. Any stories or case studies?
6. Main counterarguments?
7. How should it open?
8. Key sections (3-5)?
9. How should it close?
10. Primary audience and technical level?
11. Desired tone?
```

**Checkpoint 2**: Review interview summary, add missing details

---

### Phase 3: Content Planning & Outline

**Goal**: Create a clear structure before writing

**Process**:
1. Create outline based on:
   - Interview responses
   - Content type structure (email: intro + points + ask, blog: hook + body + conclusion)
   - Voice profile preferences (e.g., blog uses chronological flow, staff quotes)

2. Apply voice-specific patterns:
   - Email: Opening pattern, bullet points for key points, clear ask
   - Blog: Subheadings for scannability, narrative flow, reflective close
   - Social: Hook first line, call-to-action

3. Suggest length from profile:
   - Email avg: Check profile examples
   - Blog avg: Check profile examples
   - Social: Platform limits (LinkedIn 1300 chars, Twitter 280 chars)

**Email Outline Example**:
```
## Email Plan

**Subject**: [Based on purpose]

**Opening**: [Use opening pattern from profile, e.g., "I've reviewed {topic}..."]

**Body**:
- Point 1: [From interview]
- Point 2: [From interview]
- Point 3: [From interview]

**Ask**: [Specific call-to-action from interview]

**Closing**: [Use closing pattern from profile, e.g., "Let me know if..."]

**Tone**: Direct, professional, action-oriented
**Length**: ~200 words (typical for email_work)
```

**Blog Outline Example**:
```
## Blog Plan

**Title**: [Derived from central message]

**Structure**:
1. Opening: [Hook strategy - context-setting or reflective quote]
2. Background & Context
3. [Main section 1]
4. [Main section 2]
5. [Main section 3]
6. Impact & Results
7. Closing: [Forward-looking reflection]

**Voice elements**:
- Staff quotes (use throughout)
- Chronological narrative flow
- Subheadings for scannability
- Accessible but authoritative tone

**Length**: ~850 words (typical for blog_technical)
```

**Checkpoint 3**: User approves outline

---

### Phase 4: Draft Generation

**Goal**: Write content that authentically sounds like you

**Process**:
1. Load identity context:
   ```bash
   cat ~/Library/Mobile\ Documents/com~apple~CloudDocs/Geoffrey/knowledge/identity-core.json
   ```

   Use for personality alignment:
   - `communication_preferences.style` → Direct, no-nonsense, concise
   - `decision_framework` → Analytical, evidence-based
   - `telos_summary.top_3_values` → Equity, excellence, empathy

2. Apply voice characteristics as generation guidelines:
   - **Tone**: From `voice_characteristics.tone`
   - **Sentence structure**: From `voice_characteristics.sentence_structure`
   - **Vocabulary**: Common terms, avoided terms
   - **Formatting**: Bullet points, markdown, emphasis patterns
   - **Opening**: Use patterns from `opening_patterns`
   - **Closing**: Use patterns from `closing_patterns`
   - **Argument structure**: Lead with conclusion, use evidence

3. Write content naturally:
   - Use Geoffrey's native writing capabilities
   - Guided by voice profile (not rigid pattern matching)
   - Weave in interview content (examples, data, stories)
   - Match typical length from profile examples

4. Reference example emails/posts from profile for inspiration

**Generation Approach** (Hybrid):
- Write naturally with rich voice context
- Don't mechanically apply patterns
- Let voice charac