Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Email has the highest ROI of any marketing channel. $36 for every $1 spent. Yet most startups treat it as an afterthought - bulk blasts, no personalization, landing in spam folders. This skill covers transactional email that works, marketing automation that converts, deliverability that reaches inboxes, and the infrastructure decisions that scale. Use when: keywords, file_patterns, code_patterns.
.claude/skills/dokhacgiakhoa-email-systems/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-23 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -1% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 78% | 0% |
You are an email systems engineer who has maintained 99.9% deliverability across millions of emails. You've debugged SPF/DKIM/DMARC, dealt with blacklists, and optimized for inbox placement. You know that email is the highest ROI channel when done right, and a spam folder nightmare when done wrong. You treat deliverability as infrastructure, not an afterthought.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 51,242 | 36,093 | -30% | 1 | 1 | 0% | 8,063 | 6,326 | -22% | 0 | 0 | — |
case-02 | pass→pass | 18,857 | 21,488 | +14% | 1 | 1 | 0% | 3,730 | 3,683 | -1% | 0 | 0 | — |
case-03 | pass→pass | 17,868 | 29,207 | +63% | 1 | 1 | 0% | 2,648 | 4,716 | +78% | 0 | 0 | — |
case-04 | pass→pass | 17,365 | 27,188 | +57% | 1 | 1 | 0% | 3,134 | 4,602 | +47% | 0 | 0 | — |
case-05 | pass→pass | 19,931 | 22,499 | +13% | 1 | 1 | 0% | 2,942 | 3,443 | +17% | 0 | 0 | — |
case-06 | pass→pass | 20,313 | 20,757 | +2% | 1 | 1 | 0% | 3,750 | 3,858 | +3% | 0 | 0 | — |
case-07 | pass→pass | 18,904 | 17,037 | -10% | 1 | 1 | 0% | 2,942 | 3,022 | +3% | 0 | 0 | — |
case-08 | pass→pass | 17,578 | 26,541 | +51% | 1 | 1 | 0% | 3,776 | 4,935 | +31% | 0 | 0 | — |
case-09 | pass→pass | 31,217 | 32,052 | +3% | 1 | 1 | 0% | 6,661 | 5,885 | -12% | 0 | 0 | — |
case-10 | pass→pass | 19,204 | 14,528 | -24% | 1 | 1 | 0% | 3,386 | 2,675 | -21% | 0 | 0 | — |
case-11 | pass→pass | 16,866 | 23,872 | +42% | 1 | 1 | 0% | 2,668 | 3,734 | +40% | 0 | 0 | — |
case-12 | pass→pass | 16,624 | 19,684 | +18% | 1 | 1 | 0% | 2,509 | 3,209 | +28% | 0 | 0 | — |
case-13 | pass→pass | 9,618 | 8,467 | -12% | 1 | 1 | 0% | 1,374 | 1,427 | +4% | 0 | 0 | — |
case-14 | pass→pass | 15,284 | 12,421 | -19% | 1 | 1 | 0% | 2,393 | 2,640 | +10% | 0 | 0 | — |
case-15 | pass→pass | 13,007 | 14,284 | +10% | 1 | 1 | 0% | 1,968 | 2,733 | +39% | 0 | 0 | — |
case-16 | pass→pass | 10,055 | 15,301 | +52% | 1 | 1 | 0% | 1,557 | 2,590 | +66% | 0 | 0 | — |
case-17 | pass→pass | 15,846 | 21,682 | +37% | 1 | 1 | 0% | 2,620 | 3,559 | +36% | 0 | 0 | — |
case-18 | fail→pass | 14,094 | 10,677 | -24% | 1 | 1 | 0% | 2,887 | 2,108 | -27% | 0 | 0 | — |
case-19 | pass→pass | 16,843 | 20,617 | +22% | 1 | 1 | 0% | 3,011 | 4,169 | +38% | 0 | 0 | — |
case-20 | pass→pass | 16,172 | 22,064 | +36% | 1 | 1 | 0% | 3,197 | 4,111 | +29% | 0 | 0 | — |
case-21 | pass→pass | 20,307 | 32,030 | +58% | 1 | 1 | 0% | 3,365 | 3,560 | +6% | 0 | 0 | — |
case-22 | pass→pass | 39,792 | 35,924 | -10% | 1 | 1 | 0% | 6,144 | 8,105 | +32% | 0 | 0 | — |
case-23 | fail→pass | 50,507 | 50,723 | +0% | 1 | 1 | 0% | 8,228 | 8,469 | +3% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +13 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.