Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Supply Chain Bottleneck Analyzer for TikTok Shop sellers. Diagnose cash flow, inventory turnover, affiliate commissions, and return rates. Includes FBT cost analysis, influencer payout optimization, and viral product lifecycle management. No API key required for basic analysis.
.claude/skills/nexscope-ai-supply-chain-optimization-tiktok/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 61% | 0% |
Supply chain bottleneck analyzer for TikTok Shop sellers. Diagnose cash flow, inventory, affiliate costs, and return rates.
bashnpx skills add nexscope-ai/eCommerce-Skills --skill supply-chain-optimization-tiktok -g
| Feature | TikTok Shop | vs Amazon | |---------|-------------|-----------| | Fulfillment | FBT / Self-ship | FBA | | Commission | 2-8% (category) + 2% transaction | 8-15% | | Payment cycle | 7-15 days | 14 days | | Traffic source | Content-driven | Search-driven | | Return rate | Higher (impulse buying) | Medium |
Selling Price $XX
├── Product Cost
├── Inbound Shipping
├── FBT Fulfillment / Self-ship
├── Platform Fee (2%)
├── Referral Fee (2-8%)
├── Affiliate Commission (10-30%) ← TikTok-specific
├── Advertising (Spark Ads)
└── Net ProfitpythonBENCHMARKS = { "tiktok": { "gross_margin": { "healthy": 0.45, # Need to cover affiliate commission "warning": 0.35, "danger": 0.25 }, "shipping_ratio": { "healthy": 0.05, "warning": 0.08, "danger": 0.12 }, "inventory_days": { "healthy": 30, # TikTok viral cycle is short "warning": 45, "danger": 60 }, "cash_cycle": { "healthy": 60, # Fast payment "warning": 90, "danger": 120 }, "net_margin": { "healthy": 0.15, # After affiliate split "warning": 0.08, "danger": 0.03 }, # TikTok-specific metrics "return_rate": { "healthy": 0.10, # <10% healthy "warning": 0.20, "danger": 0.30 }, "affiliate_ratio": { "healthy": 0.20, # Affiliate commission ratio "warning": 0.30, "danger": 0.40 } } }
Livestream selling: 10-30% commission
Short video promotion: 10-25% commission
Top influencers: May require upfront feesTikTok return rates typically higher than traditional e-commerce (impulse buying)
Must account for:
├── Return shipping cost
├── Product damage/loss
└── Restocking fees**Sales (TikTok-specific)**
• Average Selling Price: $___
• FBT Fulfillment Fee: $___/unit
• Platform Fee: 2% (fixed)
• Referral Fee: ___%
• Affiliate Commission Rate: ___% (if applicable)
• Spark Ads Spend Ratio: ___%
**Risk (TikTok-specific)**
• Return Rate: ___%
• Return Processing Cost: $___/unitbashexport TIKTOK_APP_KEY="xxx" export TIKTOK_APP_SECRET="xxx" export TIKTOK_ACCESS_TOKEN="xxx"
| Data | API | |------|-----| | Orders | Order API | | Products | Product API | | Logistics | Logistics API | | Affiliates | Affiliate API |
TikTok Shop-specific bottlenecks:
| Item | Amazon | TikTok | |------|--------|--------| | Commission | 8-15% | 4-10% | | Affiliate split | None | 10-30% | | Payment cycle | 14 days | 7-15 days | | Return rate | 5-15% | 10-30% | | Traffic | Stable | Volatile | | Viral cycle | Long | Short |
Part of Nexscope AI — AI tools for e-commerce sellers.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 12,926 | 9,058 | -30% | 1 | 1 | 0% | 2,135 | 2,505 | +17% | 0 | 0 | — |
case-01 | fail→fail | 22,342 | 18,383 | -18% | 1 | 1 | 0% | 4,040 | 4,553 | +13% | 0 | 0 | — |
case-06 | pass→pass | 14,620 | 13,752 | -6% | 1 | 1 | 0% | 2,250 | 3,074 | +37% | 0 | 0 | — |
case-18 | pass→fail | 15,562 | 13,918 | -11% | 1 | 1 | 0% | 2,322 | 3,172 | +37% | 0 | 0 | — |
case-02 | fail→fail | 17,866 | 17,920 | +0% | 1 | 1 | 0% | 3,017 | 4,205 | +39% | 0 | 0 | — |
case-03 | fail→fail | 19,520 | 18,684 | -4% | 1 | 1 | 0% | 3,235 | 4,413 | +36% | 0 | 0 | — |
case-04 | pass→pass | 10,248 | 10,161 | -1% | 1 | 1 | 0% | 1,528 | 2,645 | +73% | 0 | 0 | — |
case-07 | fail→pass | 14,772 | 9,975 | -32% | 1 | 1 | 0% | 2,524 | 2,612 | +3% | 0 | 0 | — |
case-08 | fail→pass | 15,743 | 13,984 | -11% | 1 | 1 | 0% | 2,636 | 3,498 | +33% | 0 | 0 | — |
case-09 | fail→fail | 12,419 | 12,229 | -2% | 1 | 1 | 0% | 2,133 | 3,283 | +54% | 0 | 0 | — |
case-10 | fail→fail | 15,030 | 14,414 | -4% | 1 | 1 | 0% | 2,460 | 3,547 | +44% | 0 | 0 | — |
case-11 | fail→pass | 16,398 | 10,742 | -34% | 1 | 1 | 0% | 2,786 | 2,921 | +5% | 0 | 0 | — |
case-12 | fail→pass | 16,560 | 14,975 | -10% | 1 | 1 | 0% | 2,445 | 3,468 | +42% | 0 | 0 | — |
case-13 | fail→pass | 13,570 | 14,991 | +10% | 1 | 1 | 0% | 2,173 | 3,494 | +61% | 0 | 0 | — |
case-14 | fail→fail | 17,208 | 15,880 | -8% | 1 | 1 | 0% | 2,717 | 3,765 | +39% | 0 | 0 | — |
case-15 | pass→pass | 9,365 | 3,198 | -66% | 1 | 1 | 0% | 1,420 | 1,577 | +11% | 0 | 0 | — |
case-16 | pass→pass | 14,786 | 8,297 | -44% | 1 | 1 | 0% | 2,320 | 2,601 | +12% | 0 | 0 | — |
case-17 | fail→pass | 18,024 | 12,841 | -29% | 1 | 1 | 0% | 2,507 | 3,132 | +25% | 0 | 0 | — |
case-19 | fail→pass | 12,732 | 2,374 | -81% | 1 | 1 | 0% | 2,419 | 1,537 | -36% | 0 | 0 | — |
case-20 | pass→pass | 15,727 | 15,438 | -2% | 1 | 1 | 0% | 2,373 | 3,434 | +45% | 0 | 0 | — |
case-21 | pass→pass | 14,348 | 8,613 | -40% | 1 | 1 | 0% | 2,298 | 2,455 | +7% | 0 | 0 | — |
case-22 | pass→pass | 14,242 | 7,240 | -49% | 1 | 1 | 0% | 2,275 | 2,271 | -0% | 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 +27 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.