The platform · 8 modules
Runtime authority + signed evidence, in eight modules.
KYE Protocol™ has depth. This page is the buyer-readable view: the eight modules a procurement team needs to understand before deciding whether KYE™ belongs underneath their AI governance program. Each module is a shipped surface, not a roadmap promise.
Recent decisions
Demo dataYou learn: What KYE™ decides for each agent action, why (reason code), and who must act. You can: Replay the feed; approve or decline the held decision.
-
Accounts-payable agent pays a supplier invoice
permission_granted
Pay €412.90 to Paperhouse Supplies (demo), invoice INV-2026-0917
In-appWebhook
- accounts_payable.settle
- kye_decide
-
Procurement agent orders above its approval threshold
approval_threshold_breach
Order 40 pallets of packaging film from Kilnworks Supplies (demo), £18,400
App pushIn-appWebhook
- procurement.replenish
- kye_decide
-
Clinical summarisation agent asks to read records for marketing
permission_missing
Read 1,200 patient records to build a marketing list (demo)
In-appWebhook
- marketing.outreach
- kye_purpose_admit
-
HR operations agent acts on a revoked delegation
delegation_revoked
Change a staff member's contracted hours (demo)
In-appWebhook
- hr.contract.administer
- kye_authority_check
-
Drafting agent prepares a disclosure letter, with redaction
permission_granted
Draft a disclosure letter for matter M-2207 (demo)
In-appWebhook
- legal.matter_support
- kye_decide
-
Credit agent acts outside its jurisdiction
jurisdiction_unsupported
Approve a £4,200 credit line for an applicant in the US (demo)
In-appWebhook
- lending.adjudicate
- kye_decide
How this widget is built
- Demonstrates
- Decision feed
- Components
- Policy decision point (kye_decide)
- Decision records (kye.decision.record.v1)
- Approval routing (dual control)
- Evidence packs
- Flow
- Each agent action is decided against its chain of authority; approvals go to the delegate who holds the threshold; every outcome is sealed. Architecture diagram
- Used on
- /examples/: On the examples page, with the rest of its scenario.
- /developers/: The developer portal: what every call returns.
- /governed-ui/: GovernedUI™: the decisions it shows.
- /mcp/: The MCP server: the decisions behind its tools.
- /platform/: The platform: decisions with their chains of authority.
- /terminal/: The terminal: every decision, replayable.
The eight modules.
Each module is a shipped surface you can open today; together they give your procurement team the full authority chain, from registry to signed report.
Entity & Principal Registry
Every actor — human, agent, sub-processor, federated principal — gets a KYE™ URN. Lifecycle, classification, trust-domain. The evidence layer every other module binds to. Read more →
Authority Gateway™
The runtime decision point. Every delegated action is evaluated against the bound authority chain at the moment it's proposed — before it crosses the commit boundary. Read more →
Purpose Permission™
Every action cites a declared purpose. Every grant carries a bounded scope. Every decision is admissible against those bindings — or it doesn't reach commit. Read more →
Decision Map™
For every runtime decision: a signed structural record of inputs, rules consulted, signals fused, verdict, and obligations. The contemporaneous answer to "what did the AI decide and why".
Evidence Pack™
The canonical artefact a regulator, dispute panel or third-party auditor consumes. Replayable from the publisher's JWKS alone — no vendor dependency in the audit path. Read more →
Replay Proof™
Banking-grade replay: anyone with the publisher's JWKS can re-derive the decision offline. No vendor server in the verification path. The construction is proprietary.
Data Purpose Binding™
What data an agent touched, why it touched it, whether the purpose was valid, what evidence proves it. The KYE™ Data Governance Pack™ ships this end-to-end.
Edge Governance™
Compiled authority bundles deployable to the edge. The Compiled Authority Bundle™ + Edge Arbiter™ + Safety Floor™ + Offline Evidence Log™ ship governance into network-partitioned, regulated, or sovereign deployments.
Underneath the 8 modules
Depth is the moat. Hidden, not removed.
Behind the eight-module view, KYE Protocol™ ships 249 frameworks tracked in the live regulatory-coverage register, 10 canonical profiles, 70 rule packs, 61 sector packs, 501 dictionaries, 591 OpenAPI operations across 87+ runtime endpoints, 1165 JSON Schemas with 1449 validated examples, three SDKs (TypeScript / Python / Go), 135 conformance fixtures, a reference Gateway, embedded PDP, PEP middleware, and an MCP server. Apache 2.0 contracts; paid runtime. Banking-grade procurement teams open these on first touch.
Canonical KYE™ surfaces referenced on this page: KYE™ Edge Governance™ · KYE™ MCP Server · KYE Protocol™ · Purpose Permission.
Overview (The platform) 9 pages
- Protocol overviewTechnical overview of KYE Protocol™ — Authority Graph™, Decision Map™, taxonomy & metadata, runtime & state model, profile inventory, coverage matrix, 16…
- ValueThe added value of KYE Protocol™ across five lenses: people (end user, citizen, employee, agent, developer), value chain (buyer / vendor / merchant /…
- Whitepaper v1.0The KYE Protocol™ whitepaper: an open contract for identity, authority and audit across humans, businesses, AI agents, services, models and workflows
- Why this mattersFor every KYE Protocol™ feature: what you get without it (the regulator-grade gap), what you get with it, and which framework requirement it satisfies.…
- Customer SovereigntyKYE Protocol™ is additive and consensual. You keep your existing systems, you control exactly what you share, and KYE™ composes with your stack — it never…
- The KYE™ ThesisThe KYE™ Thesis. Why KYC, KYB, KYA and other category-based verification tools cannot govern delegated AI action
- AI Exponential → Runtime AuthorityAI capability is compounding faster than any policy cycle can respond. KYE Protocol™ moves governance to the one layer that keeps up: the execution boundary…
- The Trust Layer for Consequential ActionLegitimacy is becoming infrastructure. KYE Protocol™ is the trust layer for consequential action — the layer that decides whether an action may proceed…
How authority works 15 pages
- Action AdmissibilityKYE™ Action Admissibility Profile™ — checks whether a proposed action is admissible into the authority pipeline before any authority, formal-rule or…
- Evidence Pack™KYE™ Evidence Pack™ — the signed, replayable bundle of audit-chain entries, Decision Maps™, signatures and public-key set that a regulator or auditor…
- KYE™ Shadow Mode™KYE™ Shadow Mode™ — the runtime flag that suppresses side effects at the Commit Boundary while still running every Authority, Purpose and Decision Engine…
- Authority LifecycleThe full KYE™ authority lifecycle, end to end — from entity declaration to Authority Finality™. Ten signed states, every transition observable, every Evidence…
- Decision EngineKYE™ Decision Engine — the PDP that walks the Authority Graph™ at decision time, evaluates vocabulary-bound predicates, runs continuity + discoverability…
- Knowledge graphKYE™ knowledge graph — the live, instance-level graph of entities, authorities, delegations, capabilities, scopes, decisions and evidence
- KYE™ Agentic Governance Lifecycle™The KYE™ Agentic Governance Lifecycle™ — the ten-stage governance spine an agentic act traverses, from Intent to Renewal/Revocation, each stage bound to its…
- How KYE™ WorksThe plain-language spine: how KYE Protocol™ resolves authority before an action and preserves replayable evidence after it, in eight mechanical steps — from…
Operating model (The platform) 7 pages
- KYE™ Autonomy Tiers™KYE™ Autonomy Tiers™ — a 4-tier ladder (A0 Human-only / A1 Assist / A2 Scoped delegation / A3 Autonomous with Authority Finality™). Sharpens EC-Council ADG's…
- KYE™ Minimum Authority Controls™KYE™ Minimum Authority Controls™ (KAC-1..KAC-12) — the canonical twelve-control taxonomy KYE Protocol™ materially enforces at runtime. Mirrors EC-Council ADG's…
- Operating Model ProfileKYE™ Operating Model Profile™ — the enterprise adoption layer of KYE™. From use-case intake to readiness, authority records, authority gates, commit…
- ProfilesKYE™ has 10 canonical profiles, locked by constitution. Profiles define the conformance contract an implementation MUST support
- Execution ArchitectureThe KYE Protocol™ 10-layer execution architecture. KYE™ OWNS Authority Resolution + Evidence Finality™. KYE™ INTEGRATES with observability, model-eval, compute…
- KYE™ Consequential Actor Framework™The KYE™ Consequential Actor Framework™ is a five-level classification of how consequential an AI actor is — from L0 Informational to L4 Systemic — that…
- KYE™ Authority Studio™KYE™ Authority Studio™. Take an agent — yours or off-the-shelf — register it as a first-class principal, bind its authority, set approval policy, wire a policy…
Concepts and model 10 pages
- ConceptsThe six clickable KYE Protocol™ primitives — Identity, Authority, Scope, State, Audit, On-Behalf-Of
- Formal RulesKYE™ Formal Rules Profile™ — model permissions, obligations, prohibitions, powers, exceptions and governance meta-rules as machine-readable authority…
- Ontology ProfileKYE™ Ontology Profile™ is the semantic layer of KYE™ — shared meaning of entities, authorities, capabilities, scopes, states, decisions, evidence, profiles…
- State LibraryKYE™ State Library™ — open-source, versioned, signed reference state machines for 9 regulated industries
- TaxonomiesKYE™ taxonomies — parent / child classification of the names in the KYE™ vocabulary. Step 2 of the canonical 8-step KYE™ stack: Dictionaries → Taxonomies →…
- Continuity Module™The KYE™ Continuity Module™ checks whether the authorised meaning of an action survives delegation, interpretation, memory, context, incentives, timing…
- Continuity ProfileKYE™ Continuity Profile™ preserves the chain between intent, interpretation, delegated authority, state, decision, execution and evidence
- Entity HierarchyThe v3 KYE Protocol™ entity containment tree: Tenant siblings, Workspace siblings, and first-class Model, Tool, External App, and Audit Stream entities
Maps of KYE™ 10 pages
- Chain of AuthorityA live global view of the chain of authority: principals, delegations, agents, capabilities and grants rendered as arcs across the jurisdictions KYE Protocol™…
- The KYE™ Agent Authority Stack™An agent can restart; its authority must not drift unnoticed. The KYE™ Agent Authority Stack™ sits above any agent runtime — including open-source toolkits…
- Architecture GraphAn Obsidian-style interactive map of KYE Protocol™ — every engine, agent, rail, schema, gate and chapter as a node, every dependency as a typed edge. The same…
- The AI Value ChainThe AI value chain runs Compute → Data → Models → Agent Runtime → Workflows → Actions → Evidence → Governance → Feedback. KYE Protocol™ occupies one layer no…
- Coverage AtlasA rotating globe of where KYE Protocol™ governs: 32 jurisdictions light up with active framework coverage, 3 sit dark and requestable. Brightness is coverage…
- The Map of KYE™The Map of KYE™ renders the protocol as a cypress tree: the canopy is the chain of legitimacy — Purpose, Authority, Delegation, Execution, Evidence, Finality.…
- Gap FinderOne honest view of where KYE Protocol™ covers your regimes, and where it is a gap you can request. 32 jurisdictions covered, 3 dark and requestable. Coverage…
- Platform MapOperator 3D diagnostic of the KYE™ platform: every canonical component plotted on a rotatable dotted globe (longitude = lifecycle stage · shell = platform…
The category and its governance 12 pages
- Agentic Governance™Agentic Governance™ is the category KYE Protocol™ defines: governance FOR agentic systems — humans, AI agents, and hybrid authority — resolved to Authority…
- Authority Finality™A correct answer can still be unauthorised. Authority Finality™ is the category KYE Protocol™ defines and leads: proving every consequential AI-agent action…
- Authority Governance™Authority Governance™ (Consequential Action Governance™) is the category KYE Protocol™ defines: governing who is allowed to let an AI take a consequential…
- Governance for the AI WebGovernance for the AI Web™: reduce Systemic Authority Risk™ across the web of humans and AI agents. KYE™ Lens™, KYE™ Site Authority™, KYE™ Build-Release…
- KYE™ Governed Prompts™KYE™ Governed Prompts™ deliver Agentic Governance™ with Authority Finality™ at the prompt boundary: every prompt that drives an agent action is governed as an…
- KYE-Powered Directory™The KYE-Powered Directory™ is the public registry of organisations adopting KYE Protocol™ and Agentic Governance™. Entries appear only with explicit public…
- Open Agent Governance Foundation™The Open Agent Governance Foundation™ is the open working group standardising Agentic Governance™ — shared schemas, evidence formats, and conformance for…
- Sovereign Execution™Data sovereignty answers where AI runs. Sovereign Execution™ answers who is allowed to act — and proves it. KYE Protocol™ sits above confidential compute and…