autoresearch

1# Autoresearch Skill
2 
3Karpathy-style optimization loops for any conversion-focused content. No traffic needed. Simulated expert panel. Minutes, not weeks.
4 
5**When to use this:** Pre-launch content optimization. Generate 50+ variants, score with 5 simulated experts, evolve winners, output the best version + full experiment log.
6 
7**When NOT to use this:** Post-launch real-traffic A/B testing — that requires real analytics, not simulated scoring.
8 
9> **The sequence:** Run autoresearch FIRST to hit 85+ simulated score. Then deploy. Then validate with real traffic.
10 
11---
12 
13## What You'll Produce
14 
15Every run outputs 3 files:
16 
17| File | Purpose |
18|------|---------|
19| `{name}-optimized.{ext}` | The winning optimized content |
20| `data/{name}-experiments.json` | Full experiment log — all variants + all scores |
21| `data/{name}-optimization-report.md` | Human-readable summary with winner rationale |
22 
23---
24 
25## Expert Panel (5 Personas)
26 
27Score every variant against all 5. Batch all variants into a **single API call** per round.
28 
29| # | Persona | Scoring Lens |
30|---|---------|-------------|
31| 1 | **CMO at a mid-market B2B company (50M+ revenue)** | "Would this make me stop and engage?" |
32| 2 | **Skeptical founder** | "Do I believe this? Would I trust this company?" |
33| 3 | **Conversion rate optimizer** | "Is this clear, specific, and action-driving?" |
34| 4 | **Senior copywriter** | "Is this compelling, differentiated, and well-crafted?" |
35…