Capability: Human oversight and approval
41 pages. Search every page with Ctrl K, or open the menu.
The platform 4 pages
- The Business Case for Authority Finality™The internal-memo business case for Authority Finality™: inventories, policies and committee approvals don't prove authorised action. KYE Protocol™ proves each…
- KYE™ Agentic Authority Index™The KYE™ Agentic Authority Index™ measures whether AI agents act with valid authority, not just whether they succeed. Ten Authority Benchmark metrics, indexed…
- 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…
- The agent-protocol stackAgent frameworks (LangChain, CrewAI, OpenAI Agents SDK, Claude Agent SDK) and wire protocols (MCP, A2A, agent protocols) carry the calls. KYE Protocol™ sits…
Products 13 pages
- KYE™ GovernedUI™Review, approve, reject, evidence, and audit critical AI-agent actions before they touch production systems, customer data, money, code, or regulated…
- KYE™ GovernedUI™ (Platform and runtime)The seven OPS-suite modules of KYE™ GovernedUI™: Action Approval, Entity Passport, Authority Scope, Critical Point Review, Evidence Timeline, Approval…
- KYE™ Informed Approval™KYE™ Approval Brief™ assembles actor, principal, authority, scope, state, and risk into a structured, signed brief — so human approvers decide informed
- KYE™ Managed Authority Operations™KYE™ Managed Authority Operations™ is the productised delivery mode in which KYE Protocol™ operates your governance, authority and evidence control plane — and…
- KYE™ Mortgage Lending Action Authority Agent™KYE™ Mortgage Lending Action Authority Agent™ — a bounded, single-purpose KYE™ Governed Agent™ that govern approve/decline/terms. It is a first-class…
- KYE™ × Weights & BiasesConnect your Weights & Biases model registry to KYE™. When the model behind a production agent silently changes, regresses on an eval, or is promoted without…
- KYE™ Advanced Assurance Bundle™The KYE™ Advanced Assurance Bundle™ packages the four advanced assurance add-ons — Scenario Testing™, Approval Briefs™, Reality Coupling™, and Stable Drift…
- KYE™ Oversight Envelope™KYE™ Oversight Envelope™ closes the human-over-the-loop gap: an overseer sets an authority envelope — scope, hard limits, drift thresholds — once, then…
- Approval Briefs™Approval Briefs™ is the KYE™ GovernedUI™ add-on that turns a pending AI-agent action into a single, plain-language brief a human can approve, reject, or…
- KYE™ Agentic Lending Authority Pack™KYE™ Agentic Lending Authority Pack™ governs AI agents that make or assist consumer-credit decisions — underwriting, pricing, adverse action, arrears. Every…
- KYE™ MAS MindForge Banking Authority Pack™KYE™ MAS MindForge Banking Authority Pack™ operationalises the MAS Project MindForge AI Risk Management: Operationalisation Handbook at runtime for Singapore…
- KYE™ Shariah Structuring Authority Agent™KYE™ Shariah Structuring Authority Agent™ — a bounded, single-purpose KYE™ Governed Agent™ that govern a product structure into the Shariah Supervisory Board…
- KYE™ SME Payment Authority Pack™KYE™ SME Payment Authority Pack™ resolves, before an instruction reaches any payment rail, whether an AI payment assistant acting for an SME may cause this…
By sector 7 pages
- KYE™ AI Solutions Framework Authority Pack™The KYE™ AI Solutions Framework Authority Pack™ turns the AI Solutions Framework (an enterprise AI-adoption control framework of ~90 safeguards) into…
- KYE™ Genomics-Biosecurity Authority Pack™The Genomics-Biosecurity Authority Pack™ — a KYE™ Sector Pack Foundry™ expert-pack governing AI-generated and AI-assisted genomic sequences at the action…
- KYE™ Host-Platform Authority Adapter Pack™The KYE™ Sector Pack Foundry™ authority-adapter track for System Integrators and enterprise-AI builders. Ship a KYE™ Host-Platform Authority Adapter Pack™ that…
- KYE™ HSE Authority Pack™The KYE™ HSE Authority Pack™ governs AI agents that author, interpret or approve safety-critical HSE documents — permits-to-work, risk assessments, method…
- KYE™ ICH Authority Pack™The KYE™ ICH Authority Pack™ governs how AI-generated submission evidence under the ICH Q-series quality guidelines comes into existence at the action boundary…
- KYE™ Investment Decision Authority Pack™The KYE™ Investment Decision Authority Pack™ governs AI-assisted investment decisions and actions at the action boundary — the moment an AI-generated thesis…
- KYE™ Banking Agent Authority™Banking platforms run agentic banking; KYE Protocol™ proves whether each agentic banking action — advice, servicing, credit, payment, complaint, refund or…
Compliance, conformance & certification 2 pages
- Basel Governance HubThe Basel Governance Hub maps every BCBS 239 principle (all 14) to the KYE Protocol™ artefact, event family, and verification gate that binds it — honest…
- EU AI Act (Framework library)EU AI Act — Articles 14 + 15 (Human Oversight + Accuracy/Robustness) — control mappings to KYE Protocol™ canonical artefacts. High-risk AI systems placed on…
Solutions 2 pages
- Self-improving agentsRecursive self-improvement changes what AI can do. KYE Protocol™ governs what AI is allowed to do. When an agent's capability drifts, its prior authority is…
- KYE™ AI Adoption Navigator™The KYE™ AI Adoption Navigator™: a free, interactive pre-adoption self-assessment that maps where to use AI for the most value with the least unmanaged risk.…
Developer portal 3 pages
- KYE Protocol™ TerminalAn interactive demo of the KYE Protocol™ terminal: an agent requests an action; identity, on-behalf-of, authority, scope, state and audit are checked; the…
- Scenario Testing™ + Approval Brief™ APIKYE Protocol™ — Scenario Testing™ + Approval Brief™ API: 56 operations from scenario-testing.openapi.yaml, a published KYE Protocol™ OpenAPI contract.
- The authority tourThree tasks, about three minutes: the decision feed, a 2-of-2 approval where the initiator can never approve, and a revocation that turns the same request into…
Engage 1 page
- KYE™ Certified Agentic Delivery Pack™The KYE™ Certified Agentic Delivery Pack™ is the delivery-partner channel for consultancies, digital agencies, systems integrators, and transformation teams…
Resources 9 pages
- Action ApprovalAction Approval: Pre-action human review of an agent’s proposed action.
- Approval modesApproval modes: Locked set (§36 §6): none , single_approver , two_person , two_person_with_legal , delegated , auto (forbidden at high+ risk).
- Approval QueueApproval Queue: Multi-reviewer queue with pending / high-risk / escalations / second-approval-pending / SLA-breached / evidence-gap views.
- Authority ScopeAuthority Scope: Three-way can / cannot / needs-approval breakdown for any entity.
- Critical-action catalogueCritical-action catalogue: Twenty action classes that trigger a signed, replay-proof approval flow: send_email, send_message, delete_file, export_data…
- Decision codesDecision codes: The standard set. allow_with_constraints , require_approval , deny .
- Human involvement typesHuman involvement types: Six types. influence , direct , limit , approve , override , review .
- KYE™ Human Involvement Planner™KYE™ Human Involvement Planner™: Define where authorised humans must review, approve, limit or override AI-enabled actions across the lifecycle.
- KYE™ Human Involvement Plan™KYE™ Human Involvement Plan™: Schema-bound record of points where authorised humans must influence, direct, limit, approve, override or review an AI-enabled…