Registry / 005CapabilityProductionNo model

Price suggestion.

A deployed pricing capability that suggests a resale price from comparable items already sold. It returns nothing rather than a guess when the evidence is too thin, and a human sets every price.

Deployed capability — this configuration is running in production. Client identities are withheld.

Classified as a capability: it interprets and returns, holding no authority and routing nothing. It is a component other parts of a system build on, not an agent.

No model anywhere in this entry. It is deterministic end to end — the behaviour comes from rules, arithmetic and explicit boundaries.

Problem it stands behind

Running second-hand stock without a spreadsheetDepositors are told where their consignment stands at fixed points, automatically. Each incoming item gets a starting price from what comparable items actually sold for, and a person sets the final one.In the owner's words, on /problems

Objective

Give an operator a defensible starting number for a second-hand item, and stay silent when the comparables do not support one.

Runtime

01
Ask
An operator asks for a suggestion by brand, product type, or both.
02
Gather
Loads the price history of items already priced and sold.
03
Narrow
Matches on brand and type together, falling back to whichever is given.
04
Weigh
Takes the median of the matched prices, not the mean, so one outlier cannot move it.
05
Abstain
Below the minimum sample count it returns no figure at all, and says how few it found.

Anatomy

Context

Prices of comparable items already sold
Brand, normalised from the item title
Product type, canonicalised into groups
A configured minimum sample count

Capabilities

Canonicalise a product type
Match comparable sold items
Compute a median over the matches
Report the sample size behind a figure
Decline to suggest on insufficient evidence

Activation

Operator request for a single item

Authority

What can it change?

Autonomy: Recommend
Horizon: Operator-invoked

Read

Item price history
Brand and product-type records

Prohibited

Set a price on anything
Return a figure backed by fewer than the minimum samples
Suggest from a mean, where one outlier would carry the result

Escalate

Too few comparables — returns no figure, with the count it did find
Neither brand nor product type given — refuses the request outright

Composition and verification

Made from explicit parts.

Composition

Sold-item price history
Product-type canonicalisation
Median over matched comparables
A configured sample floor
Human pricing decision

Verification

Every figure is returned with the sample size behind it
The matching basis is named — brand, type, or both
Median rather than mean, so a single outlier cannot carry it
Abstention is the designed output, not an error path

No evidence is published for this entry. The items above are our own account of its source, read before publishing — not something you can check without taking our word for it.

Next step

What you can do with this.

This configuration runs in production, client identity withheld. To discuss the same for your operation, start with the problem rather than the system: a first call is thirty minutes and costs nothing.