Amazon Storage Fee Exposure Audit (Offline CSV)

Audit inventory-age and storage-fee exports that you already have permission to use. You supply the monthly fee schedule; the Actor calculates SKU-level storage cost, isolates the aged-stock surcharge, and highlights what needs attention.

This Actor does not access Amazon or any other website. It only analyzes the text you paste into the input. It requires no login, proxy or external API.

What it does

1. Aged-stock fee exposure per SKU – base monthly fee plus the aged surcharge tier that applies to each SKU's age.
2. Early warning – SKUs entering a higher aged tier within a configurable window, with the projected surcharge at that tier.
3. Missing data detection – flags rows with missing or unusable dimensions, units, age, or an unknown size tier, and does not guess values.
4. Prior snapshot comparison – new SKUs, SKUs removed since the earlier export, and fee changes (with a significance threshold).

Input

Field	Purpose
inventoryCsv	Current export as CSV text (comma, tab or semicolon). Requires sku. Recognized: asin, units/quantity, age_days, size_tier, length/width/height or unit_volume_cuft.
priorSnapshotCsv	Optional earlier export in the same format.
feeSchedule	Your rates: monthlyRatePerCuFt per size tier, and agedSurchargePerCuFt thresholds. No default rates are built in.
currency, dimensionUnit	Report label and in/cm.
agedFeeThreshold, approachingWindowDays, changeThresholdPct	Flagging thresholds.
maxRows	Row safety limit (default 5000, max 20000).

Sample input (the rates are placeholders, not official values; replace them with the ones from your own fee documentation):

{
  "inventoryCsv": "sku,asin,units,age_days,size_tier,length,width,height\nMUG-001,B000000001,120,95,standard,5,4,4\nLAMP-220,B000000002,40,200,oversize,30,12,10\nCASE-9,B000000004,15,60,standard,,,",
  "priorSnapshotCsv": "sku,units,age_days,size_tier,length,width,height\nLAMP-220,40,170,oversize,30,12,10",
  "feeSchedule": {
    "monthlyRatePerCuFt": { "standard": 0.8, "oversize": 0.6 },
    "agedSurchargePerCuFt": [
      { "minAgeDays": 181, "ratePerCuFt": 0.5 },
      { "minAgeDays": 365, "ratePerCuFt": 6.0 }
    ]
  },
  "currency": "USD"
}

Output

• Dataset: one item per SKU, sorted by aged surcharge, with fields such as totalVolumeCuFt, baseMonthlyFee, agedSurcharge, totalMonthlyFee, daysToNextAgedTier, changeStatus, feeDelta, severity, flags.
• Key-value store SUMMARY: totals, flag counts, top exposures, SKUs removed since the prior snapshot, warnings and caveats.

Flags: high_aged_exposure, aged_surcharge_applies, approaching_aged_tier, missing_measurements, missing_units, missing_age, unknown_size_tier, duplicate_sku, fee_increase_vs_prior, new_since_prior.

Pricing

The code emits one audit-completed event after a successful audit. Failed runs (invalid input, no usable rows) are not charged. No payable plan is configured in this repository.

Limitations

• Results are estimates from your supplied schedule, not Amazon invoices; real charges may differ because of rounding, measurement rules, unit weights, fee changes, removals or other factors this tool does not model.
• Only per-cubic-foot volume pricing is modeled. Other fee types (for example low-inventory or removal fees) are out of scope.
• Accuracy depends on your export. Rows with missing data are flagged and left unpriced instead of estimated.
• Prior-snapshot fees are recalculated with the current schedule, so changes reflect inventory shifts rather than rate changes.
• Duplicate SKUs are flagged, not merged, in the current file.
• With few SKUs (under 30) percentages and rankings are indicative only. The tool does not forecast sales or sell-through.
• Input is limited to 2 million characters per CSV and the configured row limit.

Local tests

python -m unittest discover tests (offline, standard library only apart from the Actor runtime).

Use only data you are permitted to process.