Resolve the network before assessing the risk.
Corporate networks, registry records and open-source material resolved into a structure an investigator can defend.
Risk is assessed on an entity when the risk lives in the network around it.
The same company appears under several names, numbers and addresses.
Registry extracts, filings and reporting do not join up on their own.
A conclusion without a traceable chain does not survive challenge.
A list match tells you nothing about the structure behind it.
An investigator holds the graph in a notebook.
Re-running the same investigation produces a different file.
The network resolved first, the assessment second.
Sanclarity is designed to resolve companies, people, subsidiaries, shareholders, addresses and transactions into one structure — and to distinguish an ownership link from a relationship that is only reported.
One entity, one record, separated from its near-duplicates.
The analyst reviews and challenges the output rather than manually reconstructing every exhibit. Designed to reduce analyst reconstruction — judgment, review and approval stay with the analyst.
The finding, stated in terms the record supports.
Which nodes in the network the signal rests on.
The registry extract, filing or report behind each node.
The assessment, recorded with its reasoning.
Vessels and other physical assets appear only where the underlying record supports the link. A reported association is distinguished from an ownership link.
Decision-ready, evidence-linked output.
Source → reasoning → structured output → evidence → analyst review. The intent is minimal analyst reconstruction, not an AI-generated paragraph that has to be rebuilt before it can be used.
One file per subject, assembled rather than written.
A position with the reasoning behind it, not a summary.
Intrinsic and market-based views with assumptions exposed.
Findings ranked by severity, each with supporting evidence.
Flows, counterparties and treatment set against the disclosure.
Changes that touch a stated assumption, not a notification feed.
An answer you cannot check is an opinion.
Every figure and finding keeps the disclosure it was read from, at the line level. That is what makes the output reviewable — a colleague, a committee or a regulator can open the source rather than take the conclusion on trust.
Users are responsible for reviewing and validating outputs before relying on them for material decisions.