Wedge Generator

Directly invoke when user asks:

• "Create wedges from these signals"
• "Generate messaging for [signal]"
• "Refine this wedge with A/B variants"
• "Turn this LinkedIn post into a wedge"

Inputs

Required

1. Validated Signal Data (from signal-detector)

   • Signal Type (Trust / Intent), Signal Subtype, Signal text, Prospect ICP

2. Monthly Theme (from campaign-strategist)

   • One-line monthly wedge theme

3. Weekly Wedge Strategy (from campaign-strategist)

   • A) Different angles (default) / B) Sequential story / C) A/B/C variants

4. ICP Context

   • MENA SaaS Founders / US Real Estate Brokers / MENA Beauty Clinics / US eCommerce

Optional

• Performance data (if Mode B), existing wedges (for A/B variants), geographic market

Outputs

3 Weekly Wedges (Validated)

## Weekly Wedges - [Campaign Name]

### Week 1 Wedge
**Wedge**: "[One-sentence wedge]"
**Signal Source**: [Intent/Trust: Subtype]
**ICP**: [Target ICP]
**Angle**: [Identify problem / Amplify pain / Provide solution / etc.]
**Validation**:
  - One-sentence test: PASS
  - No hedging: PASS
  - Specific (not vague): PASS
  - ICP resonance: PASS

### Week 2 Wedge
[Same format]

### Week 3 Wedge
[Same format]

## Hard Stop Rules Applied
- Rule 3 (One-sentence test): All wedges pass
- Rule 4 (Intent > Trust): Applied where both present

Wedge Creation Logic

Core Formula (from wedge-sentence-map.md)

[Observed Signal] is costing you [Specific Outcome]. [One-Line Solution].

Variations:

• Problem-focused: "[Signal] is creating [Pain]. [Solution] is the fix."
• Outcome-focused: "[Current State] generates [Low Metric]. [Better State] generates [High Metric]. That's [X]x ROI."
• Proof-focused: "[X Peers] faced [Signal]. After [Solution], they achieved [Outcome]."

For complete ICP-specific templates (MENA SaaS, US Real Estate, MENA Beauty Clinics, US eCommerce), see wedge-templates.md.

Hard Stop Rules

Rule 3: One-Sentence Test

Every wedge must pass 4 checks before output:

1. Single sentence (no multi-sentence wedges)
2. No hedging ("might", "could", "possibly", "perhaps", "maybe", "potentially")
3. No vague language ("better results", "improvement", "optimization")
4. Specificity required (must include numbers, %, $, or Xx multipliers)

See wedge-templates.md for the full validation code and PASS/FAIL examples.

Rule 4: Intent > Trust

When both Intent and Trust signals are available for a prospect, always use the Intent signal. Intent shows active pain (right now). Trust shows thought leadership (passive).

Weekly Wedge Strategies

Strategy A: Different Angles (Default)

3 wedges that each approach the monthly theme from a different lens:

• Week 1: Identify problem
• Week 2: Amplify cost
• Week 3: Provide proof

Strategy B: Sequential Story

3 wedges that build a narrative arc:

• Week 1: Story opening (the invisible loss)
• Week 2: Data reveal (the quantified gap)
• Week 3: Resolution (the solution)

Strategy C: A/B/C Variants

3 variants of the same core message with different psychology:

• Variant A: Data-driven
• Variant B: Story-driven
• Variant C: Peer-driven

Week 2 refines the winning variant. Week 3 deploys the optimized version.

Geographic Adaptation

Wedges must be adapted to the target market's communication style:

• MENA: Add relationship references, soften directness, use "we" framing, imply outcomes
• US: Direct problem statement, specific data, clear outcome, no softening
• Germany: Formal language, statistical significance, principle-first reasoning

See wedge-templates.md for adaptation examples.

Integration with Other Sub-Skills

Downstream Dependencies

After wedge-generator outputs 3 validated wedges:

asset-factory uses:

• 3 weekly wedges (exact text)
• ICP context
• Geographic market
• Channel mix
• To produce 42 assets (18 emails, 12 LinkedIn, 9 WhatsApp, 3 social)

Upstream Dependencies

1. signal-detector: Validated signals (Fit PASS, Fresh, Trust/Intent classified)
2. campaign-strategist: Monthly theme, weekly strategy, ICP context
3. wedge-sentence-map.md: ICP-specific templates

Error Handling

Conclusion

The wedge-generator sub-skill ensures:

1. Signal-specific wedges (not generic value props)
2. One-sentence clarity (Hard Stop Rule 3)
3. Intent prioritization (Hard Stop Rule 4)
4. ICP resonance (templates + language)
5. Weekly consistency (all 3 relate to monthly theme)

Output: 3 validated, one-sentence wedges ready to power 42 multi-channel assets.

Next: Pass 3 wedges to asset-factory for email, LinkedIn, WhatsApp, and social production.