A price comparison
you can question.
Find the prices worth reviewing, then inspect the evidence before treating a difference as a finding.
| Part | List price | Market median / sample quotes | List / median | Sales status | Review status |
|---|---|---|---|---|---|
| AR-101Drive belt | $84 | $30Quotes: $28 / $30 / $32 | 2.80× | Never sold | Review price |
| AR-102Bearing assembly | $145 | $140Quotes: $132 / $140 / $148 | 1.04× | Active | Compared |
| AR-103Seal kit | $96 | $40Quotes: $38 / $40 / $42 | 2.40× | Never sold | Review price |
| AR-104Mounting bracket | $52 | $18Quotes: $15 / $18 / $55 | 2.89× | Never sold | Part identity needs review |
| AR-105Filter cartridge | $42 | $46Quotes: $44 / $46 / $48 | 0.91× | Active | Compared |
| AR-106Coupling insert | $65 | —Quotes: No evidence | — | Never sold | No comparable evidence |
| AR-107Fastener pack | $90 | $10Quotes: $9 / $10 / $11 | 9.00× | Never sold | Pack quantity differs |
| AR-108Idler pulley | $115 | $74Quotes: $70 / $74 / $78 | 1.55× | Active | Compared |
A price gap is a prompt to investigate. The two qualifying rows have a combined per-unit list-minus-median difference of $110. That is neither recoverable revenue nor a savings forecast; no sales volume or cost is included.
The business question
For resale parts, how does the price on the list compare with outside evidence? The working tool brings together a read-only Kinetic data source and marketplace/search providers, retaining the evidence behind comparisons.
My contribution
I built Parts Pricing with Claude Code to answer a real pricing question. The important judgment is deciding when comparisons are credible enough to use and when a part identity or pack-size mismatch should exclude a result.
What this demonstration runs
The summary comes from the existing Python report module, using eight invented records. The quotes and confidence flags are fixtures; this page does not run marketplace searches, the provider-matching pipeline, or a live AI model. Search and filters operate on the saved report.
Inspect the example
AR-104 has a questionable part match. AR-107 compares different pack quantities. Both remain visible, but neither enters the confirmed 2×+ finding count. AR-106 has no evidence, so it receives no market price.
← All tools