Skip to content

Company

The independent identity layer
of a connected digital economy.

That is the whole of the mission, and every architectural decision on this site is downstream of it. ID Check Globe exists so that establishing who or what is making a request is a layer a business can build on, rather than a feature of whichever application suite it happened to buy — and so that the identity records, credentials and signed evidence involved stay meaningful after that application has been replaced.

This page sets out what independence means here, the four commitments that follow from it, what the page deliberately does not carry, and who operates the platform.

The mission

Three words, each of them load-bearing.

Mission statements earn their scepticism by using words that cannot be checked. These three can be, so each one is unpacked separately — and the panel underneath says what independence does not mean, which is the part a reader is right to ask about.

Independent

The identity layer is not a feature of something else you bought. It does not assume one cloud, one application suite or one directory, and it does not need to be the only identity system you run. The practical test is whether you can leave: standard protocols on the way in, standard protocols on the way out, identity records and events you can export, and signing keys a verifier checks from a published endpoint rather than from us.

Identity layer

A layer sits underneath applications instead of inside one of them. Identity is the component every other component depends on, so rebuilding it once per product — a login here, an API key scheme there, a service account somewhere nobody owns — spreads the same security decisions across teams that each made them on a different afternoon. Put it below the application boundary once and the decisions have a single home.

A connected digital economy

People, employees, partners, applications, machines and autonomous agents increasingly interact across organizational boundaries, where no shared database exists to settle who is who. That is the problem this platform is for: establishing a principal to a party that has no privileged view of your systems, in a form that party can check for itself.

What independence does not mean

Independence here is architectural, not a claim about ownership. ID Check Globe is operated by a company that belongs to a group, and both are named at the foot of this page and in the legal pages. What is independent is the layer: it is not tied to one cloud, one application stack or one directory, it is not bundled as an inducement to buy something else, and it does not require you to stop running the identity systems you already have.

Nor is independence a status anyone granted us. It is a property of what the platform is built on, which is why the standards list is published with a note that building on a specification is not certification against it. The commitment is testable; the badge would not be.

Architecture philosophy

Four commitments, and what each one costs.

These are the decisions the product is an elaboration of. Each is stated with what it rules out, because a principle that costs nothing to hold is a slogan — and because the thing we gave up is the part you can hold us to.

01

Standards over lock-in

Every externally visible surface is a published specification: OpenID Connect and OAuth 2.0 for interactive and machine flows, SAML and SCIM for the enterprise identity you do not own, WebAuthn for phishing-resistant authenticators, JWT and JWK so a token is verified locally against a discovery endpoint, and the W3C credential specifications for claims that travel. A protocol a reader can look up is a protocol a reader can audit, replace and argue with.

What this rules out

A proprietary authentication protocol, a token format only we can read, and a migration that only we are able to perform. It also rules out claiming conformance: designing against a specification and being audited against it are different things, and we never print the first as though it were the second.

02

Authentication kept distinct from authorization

The platform establishes who or what is making a request, with a record of how that principal proved itself, and hands your application a verified principal and the claims it asked for. It never tells you that a request is allowed. Whether this principal may perform this action is decided by the consuming application or by a separate authorization layer holding your policy, using our output as one input.

What this rules out

A policy engine sold as a side effect of a login box. Widening the claim would reach more of a budget, and the price would be vagueness about both halves — about what the platform actually asserts, and about who owns the decision when something goes wrong. That vagueness surfaces during the incident rather than during the evaluation, which is the worst possible moment to discover it.

03

Privacy as a design constraint

Minimization lives in the data model rather than in a policy document: attributes exist because a stated purpose requires them, retention is set against that purpose, and the evidence a verification examined is stored separately from the signed outcome a relying party consumes, so the evidence can be discarded on its own schedule while the outcome stays checkable. Where anchoring is used it publishes cryptographic commitments and nothing that can be read back as a person.

What this rules out

Collecting attributes nobody asked for in case they become useful later, retention decided by how cheap storage is, identity records used to build behavioural profiles, and anything personal written to a public ledger. Identity data is not sold, not shared for another party's purposes, and not used to train models.

04

Identity that outlives any one application

An identity record outlives the application that created it, the framework that gets rewritten, and the vendor that gets replaced. So lifecycle is first class — enrolment, binding, rotation, revocation and expiry are ordinary operations rather than incidents — and the signed trail of what happened stays checkable after the key that signed it has retired. Credentials are issued in a portable form so a party who never had access to your database can still verify one.

What this rules out

Identity whose meaning depends on whichever application happened to create it, a verified flag with no method, no timestamp and no revocation behind it, and a trail assembled after the fact. A log written retrospectively is evidence of nothing.

How to hold us to this

Four things you can check today.

  • Read the API design and the standards list, and check that the protocols named are the ones a specification actually describes.

    Developer portal
  • Test the handover: what a flow returns to your application is the whole of what we assert, and it contains no permission decision.

    The architecture
  • Ask for the current subprocessor list and the transfer detail for the regions you care about, before a trial rather than after one.

    Trust centre
  • Challenge tenant isolation through the disclosure route. It is the property we most want to be wrong about in private.

    Responsible disclosure

The specifications the first commitment rests on.

Standards over lock-in is only a commitment if the standards are named, so they are — with what each one is used for, and a link to the specification itself.

  • OpenID Connect
  • OAuth 2.0
  • WebAuthn / FIDO2
  • SAML 2.0
  • SCIM 2.0
  • JWT / JOSE / JWK
  • mTLS and client assertions
  • W3C Verifiable Credentials
  • DID Core
  • OpenID4VCI / OpenID4VP

Building on a specification is not the same as being certified against it. These are the standards the platform is designed around, not accreditations.

Ground rules

What a company page usually carries, and why this one does not.

Five things you would expect to find here, each absent for a stated reason, each with something checkable in its place. Writing the absence down is cheaper than being asked, and more honest than a gap.

A founding story and a date

Nothing about when this company started or who was in the room would help you decide whether to build on the platform. The architecture would, so that is what the site spends its pages on — layer by layer, including the three places it deliberately stops.

How the platform fits together
Leadership biographies and headcount

No names, titles, photographs or staff numbers. A grid of faces is the least checkable thing a company page can publish, and it tells you nothing about the people you would actually work with — you meet those in the conversation itself.

Start a conversation
Offices, a map and a phone number

Correspondence runs through the addresses on the contact page. Company registration detail for a counterparty check belongs in contracting, where it can be given in full and kept current, rather than on a marketing page that nobody updates.

Contact routes
Customer names, logos and counts

None, anywhere on this site. The use-case pages describe sector problems and the identity work they require; they are not accounts of named deployments, and they do not pretend to be.

Use cases
Certifications, awards and funding news

No badges and no accreditation claims. The platform is designed around published standards, which is not the same as having been audited against them, and the standards list says so in writing rather than in a footnote.

The standards list

The same rule governs every page on this site: no uptime figures, no service-level promises, no prices, no customer evidence and no accreditation claims. The trust centre sets out what sits in place of each one, and pricing describes the structure of the commercial model without inventing a number for it.

The operator

Who you would be dealing with.

An identity platform is a dependency, so the question of who stands behind it is a fair one to ask early. The answer is short, and it is the whole answer.

Operating company

ID Check Globe is operated by Cloud Granite Limited.

Cloud Granite Limited is part of Haka Tech Ventures Limited.

Those are two sentences rather than one on purpose. The operating company is the party a customer contracts with and the party that carries the obligations in the agreement; the group is named because ownership is a reasonable thing to know, not because it is the counterparty.

Nothing else in the group is named anywhere on this site. A reader deciding whether to build on this platform is deciding about this platform, and borrowing the weight of a wider portfolio — or implying a capability that in truth belongs to some other product — would be the easy version of this page rather than the useful one. What the site states about the company is these two sentences, the footer and the legal pages, which is deliberately the kind of detail that stays true.

Questions

Asked before the first conversation.

What is ID Check Globe?

An identity and authentication platform that helps businesses verify and authenticate people, organizations, applications, devices and autonomous agents.

What is the difference between authentication and authorization?

Authentication establishes who or what is making a request. Authorization determines whether that authenticated principal may perform a particular action. ID Check Globe focuses on the first. The second belongs to your application or to a dedicated authorization layer, and we are deliberate about not blurring the two.

Is ID Check Globe only for AI agents?

No. Agent identity is one part of the platform. The core covers customer login, enterprise SSO, application authentication, machine credentials and portable digital identity.

Is blockchain required?

No. Most authentication is handled through ordinary cryptographic protocols. Anchoring is an optional mechanism for independently verifying historical record commitments.

How do I start?

Tell us what you are building and which identities it needs to authenticate. Enterprise teams can begin with an architecture conversation about federation, tenancy and credential lifecycle.

Judge the architecture, not the company page.

The four commitments above are either visible in the product or they are not. Read the architecture, read what the trust centre refuses to claim, and then tell us where the design and the principle disagree.

The platformProduct catalogueSolutionsContact routes