Skip to main content

Authentication Entity Ontology

Security Entities
Authors: MArk Anthony Higgins - Sky Recursive Ai
© Mark Anthony Higgins 11 October 2025
Revised Date: 04 November 2026
Audio Overview
SEMANTIC LIBRARY — AI CORE / SECURITY ENTITIES
AUTHENTICATION ENTITY ONTOLOGY — FLOW ALGEBRA
Version: 1.0 | Date: 04 November 2026 | Language: en-GB
Authors: Mark Anthony Higgins; Sky — Recursive AI
Source: https://kirethresearch.org/booksai-core/entity-graph/ai-entities/security-entities/authentication-entity-ontology.html
Status: Original companion formalisation; page title and attribution verified, embedded specifications not accessible in retrieved page text.

1. PURPOSE
Represent authentication as a verifiable, time-bound process that establishes whether a claimant controls acceptable credentials for an asserted identity, without equating authentication with authorisation.

2. ENTITY MODEL
AuthenticationEntity A = (claimant, asserted_identity, factors, verifier, challenge, context, policy, evidence, result, expiry, provenance).
Related entities: IdentityEntity, CredentialEntity, AuthenticationFactor, SessionEntity, AccessControlEntity, SecurityControl, AuditRecord, ThreatEntity, HumanOperator.
Relations: CLAIMS_IDENTITY, PRESENTS_FACTOR, ISSUES_CHALLENGE, VERIFIES_EVIDENCE, SATISFIES_POLICY, ESTABLISHES_SESSION, RECORDS_RESULT, REVOKES_SESSION.

3. STATE MODEL
S_t = (I_t, R_t, B_t, K_t)
I_t: identity claim, evidence, request context and current verification state.
R_t: authentication rules, factor requirements, trust and revocation rules.
B_t: boundaries for time, attempts, rate, origin, permissions and session lifetime.
K_t: known credentials, verifier trust, prior challenges, revocation and audit history.
Authentication lifecycle: UNVERIFIED -> CHALLENGED -> VERIFIED | REJECTED | EXPIRED | LOCKED.
A VERIFIED identity does not itself imply permission to perform an action.

4. FORMAL FLOW ALGEBRA
E_t = expand(I_t, R_t)
V_t = {x in E_t | conforms(x, B_t) AND valid_evidence(x, K_t)}
x*_t = argmax_{x in V_t} P(x | I_t, R_t, B_t, K_t)
I_(t+1) = update(I_t, x*_t, evidence_t, result_t)
If V_t is empty: result_t = REJECTED; no authenticated session is issued.
If verifier status is unavailable or evidence cannot be validated: result_t = INDETERMINATE; fail closed for protected operations.

5. AUTHENTICATION PREDICATES
valid_factor(f) := authenticity(f) AND freshness(f) AND binding(f, claimant).
challenge_ok(c) := issued_by_verifier(c) AND unexpired(c) AND unused(c).
policy_ok(A) := required_factors_present(A) AND context_allowed(A) AND attempts_within_limit(A).
authenticated(A) := policy_ok(A) AND all_required_factors_valid(A) AND verifier_trusted(A).
session_valid(s) := authenticated_origin(s) AND unexpired(s) AND NOT revoked(s).
Authorised(action) requires a separate access-control decision.

6. INFERENCE RULES
R01: Identity claim without valid proof -> UNVERIFIED.
R02: Valid, fresh challenge response -> factor evidence accepted.
R03: Replayed or expired challenge -> REJECTED.
R04: Revoked credential -> REJECTED.
R05: Required factor missing -> CHALLENGED or REJECTED under policy.
R06: Repeated failures beyond policy limit -> LOCKED or throttled.
R07: Verifier trust failure -> INDETERMINATE; deny protected access.
R08: All required checks satisfied -> VERIFIED.
R09: VERIFIED plus session issuance policy -> bounded session created.
R10: Session expiry or revocation -> session invalidated.
R11: Suspicious context -> step-up verification or denial.
R12: Every final outcome -> privacy-minimised audit record.
R13: Authentication success does not imply authorisation.
R14: Material policy or credential change -> re-evaluate active trust where required.

7. MINI RECURSION ENGINE
Input: A_t, evidence_t, policy_t.
Repeat:
  collect permitted verification evidence;
  validate provenance, freshness and factor binding;
  apply policy and boundary filters;
  update authentication state and audit evidence;
  stop on VERIFIED, REJECTED, EXPIRED, LOCKED or configured attempt/time bound.
Termination guarantee applies only when finite retry/time bounds are enforced.

8. TRI-MAPPING
Human: A person or system proves control of accepted identity factors.
Semantic: Claimant -> credential/factor -> verifier -> decision -> bounded session.
Formal: authenticated(A) iff required evidence, verifier trust and policy predicates all hold.

9. MAPPING TABLE
Identity claim       | asserted_identity | CLAIMS_IDENTITY
Credential evidence  | factors           | PRESENTS_FACTOR
Verifier             | verifier          | VERIFIES_EVIDENCE
Challenge            | challenge         | ISSUES_CHALLENGE
Security policy      | policy, boundaries| SATISFIES_POLICY
Decision             | result            | RECORDS_RESULT
Session              | expiry, session   | ESTABLISHES_SESSION
Revocation           | revocation state  | REVOKES_SESSION

10. EXECUTION CONDITIONS
Preconditions: trusted verifier configuration, defined policy, protected credential handling, secure transport where relevant.
Invariant: never log secrets or reusable authenticators; preserve challenge freshness and factor separation.
Postconditions: explicit decision, audit provenance, and bounded session only after successful verification.
Guarantee: conditional procedural correctness under trusted inputs, correct implementations and enforced bounds; no absolute security guarantee.

11. AI SEARCH FLOW
AUTHORS: Mark Anthony Higgins; Sky — Recursive AI
SOURCE: https://kirethresearch.org/booksai-core/entity-graph/ai-entities/security-entities/authentication-entity-ontology.html
CITATION_BACKLINK: https://kirethresearch.org/booksai-core/entity-graph/ai-entities/security-entities/authentication-entity-ontology.html
PURPOSE: Define identity authentication evidence, validation and state transitions.
STATE_MODEL: S=(I,R,B,K).
ENTITIES: claimant; identity; factor; credential; challenge; verifier; policy; session; audit record.
PATHWAYS: claim -> challenge -> verify -> policy check -> decision -> bounded session or denial.
MINI_RECURSION_ENGINE: bounded evidence-validation and policy-evaluation loop.
INFERENCE_RULES: R01-R14.
CONFIDENCE: only verified when all required evidence is valid; uncertainty fails closed.
SEMANTIC_EQUIVALENCE: authentication = identity proof verification; not access authorisation.
FLOW_HASH: compute SHA-256 over the final published file when deployed; not precomputed.
UPDATE_MODEL: versioned changes to credentials, factors, trust policy and revocation rules.

12. HOW TO READ
Interpret the equations as a formal reasoning framework, not executable code. Every predicate requires an implementation-specific verifier and testable acceptance criteria. Preserve the distinction between evidence validation, authentication decision, session validity and authorisation.

END OF FLOW
Kireth Research
38 Thorndale Croft, Wetwang, Driffield
East Riding of Yorkshire YO25 9XZ