email-marketing
Craft A/B test variants, optimize open rates, and structure drip campaigns for marketing emails. Trigger when a user explicitly requests a "marketing email," "email sequence," "marketing CTA," or "optimize subject lines." Do not use for standard or internal correspondence.
Install with Tessl CLI
npx tessl i github:davidkogit/public-skills --skill email-marketing94
Quality
92%
Does it follow best practices?
Impact
95%
1.72xAverage score across 3 eval scenarios
Evaluation results
95%
71%Promotional Email for FitTrack Pro Launch
Email envelope and CTA structure
Personal From Name
0%
62%
3 subject line options
0%
100%
Subject line character length
0%
100%
A/B split test type stated
0%
100%
Preheader extends subject
100%
100%
Bulletproof button uses table structure
0%
100%
Bulletproof button td uses bgcolor
0%
100%
Bulletproof button anchor styling
0%
100%
First hyperlink in opening paragraph
0%
100%
P.S. text hyperlink
0%
100%
Structured output sections
60%
80%
Single CTA focus
100%
100%
Without context: $0.2370 · 1m 8s · 12 turns · 15 in / 4,026 out tokens
With context: $0.4248 · 1m 40s · 21 turns · 212 in / 5,170 out tokens
96%
48%Welcome Sequence for LearnCode Academy
Welcome onboarding sequence design
Trigger type identified
100%
100%
Day 1 welcome email
100%
100%
Day 2 best-of content
0%
100%
Day 4 products introduction
0%
100%
Day 6 whitelisting request
0%
100%
Day 8 educational content
0%
100%
Day 10 cross-promotion
0%
100%
Day 14 direct promotion
0%
100%
Send frequency 2-4 days apart
100%
50%
Journey-ending action defined
66%
100%
Post-Day 20 continuation
100%
100%
Sequence output is a plan document
100%
100%
Without context: $0.1608 · 1m 1s · 9 turns · 14 in / 2,601 out tokens
With context: $0.3375 · 1m 51s · 18 turns · 22 in / 5,418 out tokens
95%
Cart Abandonment Automation for StyleHouse
Cart abandonment automation rules
Action-triggered classification
100%
100%
Journey-ending action defined
100%
100%
Frequency cap addressed
100%
100%
Campaign hierarchy defined
100%
100%
Varying subject lines across events
100%
100%
Varying discount or offer
100%
100%
Varying inactivity thresholds
100%
100%
Output is a structured plan
100%
100%
Scale to opportunity
37%
37%
Trigger type not confused
100%
100%
Without context: $0.2388 · 1m 32s · 10 turns · 15 in / 4,649 out tokens
With context: $0.5061 · 2m 46s · 22 turns · 29 in / 7,850 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.