# Layerup

- [Platform Overview](https://docs.uselayerup.com/overview.md): Layerup is the Agentic AI Operating System for Insurers — a decision-centric operating substrate connecting data, logic, models, action surfaces and security into a single governed plane.
- [Architectural Principles](https://docs.uselayerup.com/principles.md): Ten invariants the platform enforces by construction. Every section is an implementation of one or more of them.
- [Platform Topology](https://docs.uselayerup.com/topology.md): Layerup organised as eight horizontally cooperating planes plus three cross-cutting services. Each plane has a contract; planes communicate only across those contracts.
- [Core Concepts & Vocabulary](https://docs.uselayerup.com/concepts.md): Every term defined here is used throughout the documentation with the exact meaning given. Sections 5–22 assume this vocabulary.
- [The Insurance Ontology](https://docs.uselayerup.com/ontology/objects.md): The canonical, version-controlled object model. 28 objects across seven groups: Parties, Master data, Underwriting, Claims, Workflow, Evidence, and Governance.
- [Ontology Evolution, Branching & Versioning](https://docs.uselayerup.com/ontology/evolution.md): How the Insurance Ontology evolves without breaking active runs, queues, integrations, or the audit chain. Semver rules, deprecation, branching, pinning, compatibility tests, and replay.
- [Data Plane — Ingestion](https://docs.uselayerup.com/data/ingestion.md): The Data Plane accepts heterogeneous signal from many channels, classifies, hashes, normalises, and presents it as a typed, content-addressed stream of objects.
- [Data Abstraction, Mapping, Lineage & Replay](https://docs.uselayerup.com/data/lineage.md): The Data Abstraction Layer: where heterogeneous input becomes typed, audited, replayable Ontology objects with full provenance, lineage graph, and governed RAG retrieval.
- [Tool Pattern Library](https://docs.uselayerup.com/tool-patterns.md): The Logic Plane exposes capability only as tools: typed, versioned, idempotent, permissioned, audited, eval-gated functions. Ten patterns cover every capability.
- [Agent Runtime Architecture](https://docs.uselayerup.com/runtime/runtime.md): The Reasoning Plane: a multi-tenant, sandboxed, queue-driven runtime that turns typed inputs into typed Decisions and proposed Actions under permission, budget and time bounds.
- [Planner, Executor, Verifier & Memory](https://docs.uselayerup.com/runtime/internals.md): Each agent run is a coordinated loop of four components. Planner produces a structured plan; executor dispatches it; verifier rules a candidate Decision; memory holds run-scoped state.
- [Model Gateway](https://docs.uselayerup.com/models/gateway.md): The only path from the Reasoning Plane to a model. Vendor-neutral routing across capability lanes, approved model registry, asset separation, region pinning, and a six-stage Model Fabric.
- [Evaluation, Benchmarks & Drift](https://docs.uselayerup.com/models/evaluation.md): Promotion of any model, prompt, tool, agent, or ontology change is gated by the evaluation harness. Evaluation is continuous; drift is a first-class signal that demotes assets automatically.
- [Action Plane](https://docs.uselayerup.com/action-plane.md): The only path from a Decision to a mutation of a system of record. Every effect is typed, idempotent, approved, audited, and reversible. The threshold between AI suggestion and governed agency.
- [Identity, Access & Safety](https://docs.uselayerup.com/security/identity.md): Compliance & Control Layer: identity, RBAC/ABAC, markings, purpose-tagged access, break-glass, encryption, and the Kill Switch that lets operators stop any AI work at any granularity in seconds.
- [Tool Invocation Governance](https://docs.uselayerup.com/security/tool-governance.md): Every tool call is mediated by the policy decision point (PDP). The PDP is small, deterministic, and audited. Its rulings are the gating contract between intent and effect.
- [Audit, Decision Lineage & Action Lineage](https://docs.uselayerup.com/security/audit.md): Tamper-evident by construction: every governance-relevant event is appended to a hash-chained log per tenant. Decision and Action lineage are graph projections over the audit chain.
- [Observability & Telemetry](https://docs.uselayerup.com/operations/observability.md): The substrate is observable end-to-end. Every request, plan step, model call, PDP ruling, and action transition emits typed spans, metrics, and logs. OTel-compatible throughout.
- [Deployment Topology & Tenancy](https://docs.uselayerup.com/operations/deployment.md): Three primary topologies: tenant-private cloud, on-prem/hybrid, and sovereign cloud. Each preserves the same architectural contract. Multi-region federation via regional cells.
- [Reliability, Throughput & SLOs](https://docs.uselayerup.com/operations/reliability.md): Queue-driven, backpressured, and circuit-broken. SLOs declared per surface; capacity planning per (tenant, lane, adapter) tuple. The platform never silently drops work.
- [Versioning, Release Gates & Failure Modes](https://docs.uselayerup.com/operations/release-gates.md): Releases are configuration promotions across a fixed environment ladder, gated by evals, drift, and error budgets. Every plane has a documented failure-mode matrix; failure is a first-class state.
- [Enterprise Scale & Federation](https://docs.uselayerup.com/enterprise-scale.md): For the global CTO: how the substrate operates across geography, LOB, workflow, and systems at enterprise scale — all on one substrate, with sub-linear marginal cost per addition.
- [Reference Appendix](https://docs.uselayerup.com/appendix.md): Reference material for technical diligence: glossary, object index, tool-pattern index, action registry, role × permission matrix, event taxonomy, and integration / rollout checklists.
- [Agents Overview](https://docs.uselayerup.com/agents/overview.md): Layerup AI Agents are sovereign, containerized reasoning workloads that deploy entirely within your private cloud — no SaaS boundary, no shared infrastructure, no egress to Layerup.
- [Architectural Principles](https://docs.uselayerup.com/agents/principles.md): Seven non-negotiable invariants that govern every Layerup AI Agent deployment. These are structural properties of the architecture — not configuration options.
- [Deployment Options](https://docs.uselayerup.com/agents/deployment/options.md): Choose how Layerup AI Agents are hosted and operated: on-premises in your own data center, in your private cloud account, or in Layerup's dedicated HIPAA-compliant cloud. All three are legally permissible under HIPAA.
- [Native VPC Deployment Paradigm](https://docs.uselayerup.com/agents/deployment/paradigm.md): The foundational deployment model: Layerup AI Agents run entirely within your private cloud network boundary, procured via cloud marketplace, isolated by construction from all external infrastructure.
- [Infrastructure & Compute Runtimes](https://docs.uselayerup.com/agents/deployment/compute.md): Container image delivery, private registry integration, pre-deployment security scanning, execution environment options, and the compute scaling model for Layerup AI Agents.
- [Agent Core Service Suite](https://docs.uselayerup.com/agents/deployment/agent-core-services.md): A breakdown of every Amazon Bedrock AgentCore service, which ones a Layerup deployment activates by default, which are optional, and what your cloud governance team needs to approve before go-live.
- [Companion Dashboard](https://docs.uselayerup.com/agents/deployment/dashboard.md): The Layerup companion dashboard is an optional second container that provides a browser-based UI for manual case submission and structured output review. It deploys inside your VPC alongside the agent and adds no new egress paths.
- [Layerup's Cloud](https://docs.uselayerup.com/agents/deployment/layerup-cloud.md): Network topology, connectivity options, API integration mechanism, dedicated cluster infrastructure, and data residency specifics for Option 3: Layerup-hosted dedicated cloud.
- [Multi-Line Deployment](https://docs.uselayerup.com/agents/deployment/multi-line.md): How to deploy Layerup AI Agents across multiple lines of business — IDI, Life, Group Disability, and others — within a single deployment, with complete per-line data isolation, independent AOP configuration, and shared compute efficiency.
- [Identity, Access & AI Guardrails](https://docs.uselayerup.com/agents/security/iam.md): Least-privilege IAM model for AWS and Azure, native AI guardrail configuration, and the principle that the agent introduces no new security surface — it operates within the perimeter you already own.
- [Encryption & Supply Chain Security](https://docs.uselayerup.com/agents/security/data-protection.md): Encryption standards across every data layer, container image supply chain provenance, SBOM disclosure, CVE remediation SLAs, penetration testing, and SOC 2 Type II coverage.
- [SSO & API Authentication](https://docs.uselayerup.com/agents/security/sso.md): Single Sign-On configuration for the companion dashboard and OAuth 2.0 client credentials for system-to-system API authentication — covering both Option 2 (private cloud) and Option 3 (Layerup's Cloud).
- [Data Pipeline & Ingestion](https://docs.uselayerup.com/agents/data/pipeline.md): End-to-end data flow architecture, event-driven ingestion triggers, multi-document context assembly, OCR pipeline, third-party feed integration, and ambiguity flagging — all within your network boundary.
- [Output Schema & Write-Back](https://docs.uselayerup.com/agents/data/output.md): The structured JSON output payload produced by Layerup AI Agents — nine top-level objects, fully machine-readable — and the write-back integration patterns for delivering it to your systems of record.
- [Observability, Logging & Audit Trails](https://docs.uselayerup.com/agents/observability/telemetry.md): Telemetry architecture across CloudWatch and Azure Monitor, step-level reasoning capture, deterministic audit trails for regulatory examination, log retention and immutability, and long-term archival.
- [Error Handling & Resilience](https://docs.uselayerup.com/agents/observability/resilience.md): The Layerup agent never silently fails. Every error condition — OCR failure, model timeout, guardrail intervention, low confidence — results in a deterministic, logged fallback action and human escalation.
- [Agent Operating Procedure & Configuration](https://docs.uselayerup.com/agents/lifecycle/aop.md): The AOP is the agent's operating contract — a version-controlled, human-readable configuration artifact that encodes your underwriting rules. How it is authored, validated, updated, and governed.
- [CI/CD Pipeline & Deployment Strategies](https://docs.uselayerup.com/agents/lifecycle/cicd.md): Layerup's release process, your container image promotion pipeline, blue/green deployment, shadow mode validation, and the progressive volume rollout model from POC to full deployment.
- [Agent Update Lifecycle](https://docs.uselayerup.com/agents/lifecycle/updates.md): How Layerup manages agent updates to dedicated SaaS clusters: advance notice windows, maintenance schedules, freeze request policy, pre-update validation, and rollback SLA.
- [Confidence Engine](https://docs.uselayerup.com/agents/confidence/engine.md): The confidence engine is the backbone of every Layerup AI Agent recommendation — a multi-signal, evidence-grounded scoring system that governs whether a case is resolved, flagged, or escalated to a human underwriter.
- [Deployment Readiness Reference](https://docs.uselayerup.com/agents/appendix.md): Complete AWS and Azure deployment readiness checklists, owner assignments for each prerequisite, and key contacts for the Layerup implementation engagement.
- [Embedding the Chat Widget](https://docs.uselayerup.com/integrations/chat-widget/embed.md): Add a fully managed support chat widget to any website or mobile app with a single script tag.
- [Identity Verification](https://docs.uselayerup.com/integrations/chat-widget/identity-verification.md): Pass a server-side HMAC hash to identify authenticated users and skip redundant OTP prompts.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.