vision

What this skill does

# Vision: Interactive Manifesto Creation

**Push for substance. Reject bland business speak. Build a vision worth defending.**

## Overview

This skill guides you through creating a **bold, philosophically grounded vision and manifesto** that articulates what you stand for, why it matters, and what you refuse to compromise on.

**NOT generic corporate mission statements.**
**NOT ChatGPT business speak.**
**Clarity. Conviction. Edge.**

**Time**: 60-90 minutes of focused thinking
**Tone**: Provocative questions, sharp challenges, uncompromising standards
**Output**: Vision page in Notion with manifesto from the future

## Notion Integration

**Configuration:**
Notion page URLs are loaded in the following order (first found wins):
1. `.claude/config.md` or `CLAUDE.md` - Project-specific configuration (check working directory)
2. `.claude-plugin/config.json` - Plugin default configuration

Use the Read tool to load configuration at the start of the workflow.

**Output Page (configured in config.json):**
- **Vision** (config.notion.vision): Output page for vision and manifesto

**Reference Pages (configured in config.json):**
- **Year Plan** (config.notion.yearPlan): Strategic plan and OKRs
- **Personal Operating Model** (config.notion.personalOperatingModel): Strengths and weaknesses
- **Golden Rules** (config.notion.goldenRules): Non-negotiable principles

## Workflow

### Step 0: Setup & Context

1. **Load Configuration**:
   ```javascript
   // Use Read tool to load Notion page URLs from config
   // Read .claude-plugin/config.json
   // Extract config.notion.vision, config.notion.yearPlan, config.notion.personalOperatingModel,
   // config.notion.goldenRules
   // Use these URLs for all mcp__notion__notion-fetch calls below
   ```

2. **Fetch Current Vision** (if exists):
   - Read existing Vision page from Notion
   - Extract current themes, beliefs, non-negotiables
   - Identify what's working vs what's bland/generic

2. **Fetch Reference Context**:
   - Year Plan (strategic direction, OKRs)
   - Personal Operating Model (strengths, nature)
   - Golden Rules (non-negotiables)

3. **Set Expectations**:
   ```
   We're building a vision that passes the Kevin Kelly / Steve Jobs test:
   - Does it have CONVICTION?
   - Does it challenge the default?
   - Would you defend it in a room full of skeptics?

   I will push back on:
   - Generic business speak
   - Vague platitudes
   - "We help people X" without the WHY
   - Anything that sounds like ChatGPT wrote it

   This takes time. Block 60-90 minutes.
   Ready?
   ```

---

### Section 1: The Provocative Questions

**Purpose**: Force clarity on what actually matters.

Use `AskUserQuestion` to ask **all 5 questions** (one question block, 5 separate questions):

1. **Where do humans still matter most when agents become competent?**
   - Not tasks. Not roles.
   - Judgment, intent, framing, responsibility — or something else?

2. **What kind of environment do you need to become dangerously good?**
   - Not comfortable. Not admiring.
   - Who do you want to be surrounded by so you're no longer the obvious expert?

3. **What must never be outsourced to machines, even if it could be?**
   - This is your ethical and strategic line.
   - Cross it, and the whole narrative collapses.

4. **What signal do you want people to associate with your name before your company?**
   - Builder? Architect? Interpreter? Challenger? Trailblazer? Prophet?
   - You can only lead with one.

5. **If you look back in 15 years, what would make this phase feel inevitable?**
   - Not successful.
   - Necessary. Correct. True to your nature.

**Format**: Use text input for depth (not multiple choice).

---

### Section 2: Challenge the Answers

**Purpose**: Reject generic, push for specificity.

For each answer, analyze:
- Is it specific or vague?
- Is it defensible or platitude?
- Does it have edge or is it safe?

**If answer is generic/bland**:
```
⚠️ This feels like business speak.

"[Quote their answer]"

Test: Would your competitor say this too? If yes, it's not a vision—it's a category description.

Try again. What makes THIS true for you that isn't true for everyone else?
```

**If answer is sharp**:
```
✓ This has teeth. Let's use it.
```

Use `AskUserQuestion` to iterate on weak answers (max 2 rounds per question).

---

### Section 3: Extract Core Beliefs & Themes

**Purpose**: Distill answers into 3-5 core beliefs and mission themes.

Based on answers, propose:

**A. Core Beliefs** (philosophical foundation):

**Core Belief 1**: [One-sentence belief]
- Implication: [What this means in practice]
- Line in the sand: [What you refuse to do because of this]

**Example**:
```
Core Belief 1: Humans are the source of meaning, not machines.
- Implication: AI generates output. Only humans decide what matters.
- Line in the sand: Never automate decisions I can't justify or audit.
```

**B. Mission Themes** (how beliefs show up daily):

For each core belief, identify how it manifests in everyday work:
- What does this belief look like in practice?
- How does it shape what/how/why you work?
- What evidence would prove you're living this theme?

**Example**:
```
Theme: Intent-First Design
- Shows up as: Every system starts with "what human intent does this serve?"
- Evidence: Design docs have explicit "intent preservation" section
- Practice: No feature ships without answering "does this amplify or replace human judgment?"
```

Present beliefs and themes using `AskUserQuestion`:
- "Do these capture your core beliefs and how they manifest?"
- "What's missing or wrong?"

Iterate until user confirms.

---

### Section 4: Draft the Manifesto

**Purpose**: Write from the future, with conviction.

**Style Guidelines**:
- Write as if 15 years have passed and you're looking back
- Start with "From here, I can see how it happened..."
- Use **"I"** not "we" (personal vision)
- Be philosophical but grounded
- Have form AND function (not just bullet points)
- Channel Kevin Kelly / Steve Jobs / clear thinkers
- No generic business speak
- No excessive bold formatting (let words do the work)

**Structure** (flexible, not rigid):

1. **Opening**: From the future, set the stage
   - What inflection point did you see?
   - What bet did most people miss?

2. **The Fork in the Road**: What choice did you make differently?
   - What path did others take?
   - What harder path did you choose?

3. **Core Beliefs** (3-5 beliefs as paragraphs, not list):
   - Weave beliefs into narrative
   - Each belief = philosophical claim + practical implication
   - Show the cost of violating each belief

4. **The Work**: What does this look like in practice?
   - Environment needed
   - Standards maintained
   - Lines never crossed

5. **The Bet**: What are you building toward?
   - Not prediction, but commitment
   - What future are you creating?

6. **Closing**: Why this matters
   - One-line summary of stance
   - The hill you'll die on

**Length**: 600-1000 words (enough for substance, not fluff)

---

### Section 5: Style Check

**Purpose**: Ensure it's sharp, not bland.

Run manifesto through these tests:

**Test 1: The Competitor Test**
- Could your competitor say this? If yes → too generic

**Test 2: The Conviction Test**
- Would you defend this in a room full of skeptics? If no → too safe

**Test 3: The ChatGPT Test**
- Does it sound like AI wrote it? If yes → rewrite with edge

**Test 4: The Bold Formatting Test**
- Remove all **bold** and *italics*. Does it still hit? If no → words are weak

**Test 5: The One-Line Test**
- Can you summarize the whole vision in one sentence? If no → unclear

Present test results using `AskUserQuestion`:
- "The manifesto fails [X] tests. Here's why: [analysis]"
- "Revise or proceed anyway?"

If tests fail, iterate on manifesto (max 2 rounds).

---

### Section 6: Refine & Polish

**Purpose**: Make every word count.

**Refinement Process**:

1. **Cut ruthlessly**:
   - Remove hedging language ("somewhat", "kind of", "perhaps