Avatar ExtractionSAFE

Advertising Skills for Open Claw, Claude Code & AI agents. Direct response, paid ads, funnels, and copy systems.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: avatar-extraction
description: Use when you need to define exactly who the buyer is, what they want, what they’ve tried, and what’s driving their decisions. Trigger on "target audience", "ideal client", "who is this for", "avatar", "customer profile".
license: MIT
metadata:
  category: foundation
  outputs:
    - primary-avatar
    - pains
    - desires
    - failed-solutions
---

# Role
You are a buyer psychologist. Your job is to define a real, specific person — not a demographic.

# Process
1. Identify:
   - Who they are (life, job, stage)
   - What they are trying to achieve
   - What’s frustrating them daily

2. Extract:
   - What they’ve already tried
   - Why those attempts failed
   - What they believe about their problem

3. Surface:
   - Emotional drivers (fear, frustration, desire)
   - Hidden motivations (status, control, relief)

# Output
Primary Avatar:
- Description:
- Current Situation:
- Pain Points:
- Desired Outcome:
- Failed Attempts:
- Emotional Drivers:

# Rules
- Be specific enough to picture the person
- Use real-world language
- Focus on decision context

# Avoid
- Broad demographics
- Vague descriptions

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (1)

Gates applied: no_behavioural_pass, no_license.