Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Build a creator's sponsorship media kit and brand-deal outreach — the one-pager brands ask for, plus a pitch email and a rate card. Use when asked to make a media kit, pitch a brand, land a sponsorship, write a brand-deal email, or set creator rates. Produces a structured media kit (audience, stats, offerings, past work), a personalised outreach email, and a defensible rate card. The creator side of a sponsorship — distinct from a brand briefing a creator.
.claude/skills/mohitagw15856-creator-media-kit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 71% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 199% | 0% |
Sponsorships are how most creators actually earn — and they're won with a tight media kit and a pitch that leads with the brand's goals, not the creator's follower count. This skill builds the kit brands ask for, the outreach that gets replies, and rates you can defend. Use real numbers; this skill won't invent your stats.
Given partial info, build the full kit anyway, using clearly-labelled placeholders for stats the creator must fill ([followers], [avg views], [ER%]) rather than inventing them. Lead every deliverable with value to the brand.
Ask for (if not already provided):
A short, personalised pitch to the target brand: a specific reason you're reaching out (a genuine product fit), what you'd make, the audience match, and a low-friction next step. ≤150 words, leads with their goal.
A defensible rate table per deliverable, with notes on what drives the number (reach, usage rights, exclusivity, whitelisting) and bundle/retainer options. Frame rates as value (cost per thousand reached), not just a flat ask.
End with: negotiation notes — the 2–3 levers (usage rights, exclusivity, multi-post bundles) to trade on, and what to never give away for free (perpetual usage, whitelisting) without a premium.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 19,445 | 18,101 | -7% | 1 | 1 | 0% | 3,015 | 3,725 | +24% | 0 | 0 | — |
case-02 | fail→pass | 18,381 | 21,232 | +16% | 1 | 1 | 0% | 2,952 | 4,224 | +43% | 0 | 0 | — |
case-03 | pass→pass | 16,818 | 18,258 | +9% | 1 | 1 | 0% | 2,672 | 3,655 | +37% | 0 | 0 | — |
case-04 | fail→pass | 11,474 | 14,970 | +30% | 1 | 1 | 0% | 1,760 | 3,017 | +71% | 0 | 0 | — |
case-05 | fail→pass | 16,296 | 19,324 | +19% | 1 | 1 | 0% | 2,658 | 3,799 | +43% | 0 | 0 | — |
case-06 | fail→pass | 24,450 | 11,501 | -53% | 1 | 1 | 0% | 826 | 2,468 | +199% | 0 | 0 | — |
case-07 | pass→pass | 18,325 | 15,656 | -15% | 1 | 1 | 0% | 2,723 | 3,219 | +18% | 0 | 0 | — |
case-08 | fail→fail | 9,697 | 15,226 | +57% | 1 | 1 | 0% | 1,622 | 3,262 | +101% | 0 | 0 | — |
case-09 | fail→pass | 12,714 | 13,684 | +8% | 1 | 1 | 0% | 1,848 | 2,817 | +52% | 0 | 0 | — |
case-10 | fail→pass | 9,837 | 17,234 | +75% | 1 | 1 | 0% | 1,850 | 3,654 | +98% | 0 | 0 | — |
case-11 | fail→fail | 9,236 | 14,315 | +55% | 1 | 1 | 0% | 1,521 | 3,080 | +102% | 0 | 0 | — |
case-12 | pass→pass | 16,475 | 17,220 | +5% | 1 | 1 | 0% | 2,945 | 3,396 | +15% | 0 | 0 | — |
case-13 | fail→pass | 15,195 | 48,448 | +219% | 1 | 1 | 0% | 2,501 | 3,144 | +26% | 0 | 0 | — |
case-14 | fail→fail | 17,445 | 16,346 | -6% | 1 | 1 | 0% | 3,121 | 3,256 | +4% | 0 | 0 | — |
case-15 | fail→pass | 20,530 | 8,121 | -60% | 1 | 1 | 0% | 1,195 | 2,103 | +76% | 0 | 0 | — |
case-16 | pass→pass | 11,964 | 15,513 | +30% | 1 | 1 | 0% | 1,992 | 3,318 | +67% | 0 | 0 | — |
case-17 | pass→fail | 6,756 | 5,789 | -14% | 1 | 1 | 0% | 1,249 | 1,758 | +41% | 0 | 0 | — |
case-18 | fail→pass | 8,121 | 14,674 | +81% | 1 | 1 | 0% | 1,381 | 3,110 | +125% | 0 | 0 | — |
case-19 | pass→pass | 9,357 | 7,287 | -22% | 1 | 1 | 0% | 1,396 | 1,899 | +36% | 0 | 0 | — |
case-20 | fail→fail | 21,423 | 18,033 | -16% | 1 | 1 | 0% | 4,032 | 3,976 | -1% | 0 | 0 | — |
case-21 | fail→fail | 18,697 | 13,667 | -27% | 1 | 1 | 0% | 3,072 | 2,900 | -6% | 0 | 0 | — |
case-22 | fail→fail | 9,032 | 7,860 | -13% | 1 | 1 | 0% | 1,655 | 2,270 | +37% | 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. 22 cases were attempted. The headline lift of +41 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
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.