Does this store deliver and refund? Measured with carriers and payment processors.
Scored across 3 tools
Each tool has a clearly distinct role: find_stores lists and summarizes stores, get_store retrieves one full record by domain, and how_holdmark_works explains methodology. There is no overlap that would cause misselection.
find_stores and get_store follow a consistent verb_noun pattern, but how_holdmark_works is a noun-phrase/meta-tool name that breaks the pattern. The deviation is minor and still readable.
Three tools are well-scoped for a read-only store-facts registry: list, detail, and methodology. Each earns its place, and no unnecessary tools are present.
The surface covers the full lifecycle for this read-only domain: discovering stores, reading detailed records, and understanding how figures are calculated and what is not measured. No create/update/delete operations are needed because the registry is externally populated.
List the online stores in the holdmark registry, with a summary of their published facts. Each store publishes the same facts under the same rules: whether it delivers its orders (checked with the carrier) and whether it refunds money (checked with the payment processor). holdmark does not rank stores and has no score, so compare them yourself by whatever the person actually cares about. The registry only contains stores that connected their data. A store missing from it has not been judged.
Inputs: sort, limit, order, offset, +2 more.
Read the full holdmark record for one store, by its domain. Figures are rounded against the merchant, order volume is published in bands rather than exact numbers, and every figure carries the period it covers and the source it came from. A fact with no data answers "not_measured" with its reason, never "no". If the domain is not in the registry you get a plain answer saying so, which is not a verdict about that store.
Inputs: domain (required).
How every figure in the registry is calculated, what the words mean, and above all what holdmark does NOT measure. Read this before drawing a conclusion from any record: it is what stops a figure being read as something it is not.
Inputs: no named parameters.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Source-dated facts on online shops for AI agents and buyers: returns, shipping, payment, trust.
Fail-closed refund, cancellation, return, and trial checks for 100 US subscription vendors.
Verified deals, store policies & a trust score for thousands of online retailers. No auth.
Is a website ready for AI shopping agents? Readiness score (0-100) + agent shopping simulation.
Add one secure layer between your agents and this server.
