Pattern Library
AI Governance Pattern Library
138 patterns across 16 domains. Filter by regulatory framework to find the control that satisfies a specific obligation.
Agent Checkpoint and Recovery
The Agent Checkpoint and Recovery Pattern defines the durable execution architecture that…
EAAPL-AGT010 — AI Agent Cost Governance Architecture
Agentic AI systems that autonomously execute multi-step workflows—calling tools, querying APIs, writing…
Agent Identity and Authorisation
The Agent Identity and Authorisation Pattern defines the security architecture governing who…
Agent Sandboxing
The Agent Sandboxing Pattern defines the secure execution environment that isolates AI…
Agent Tool Registry
The Agent Tool Registry Pattern defines the centralised catalogue and invocation framework…
Event-Driven Agent
The Event-Driven Agent Pattern defines the architecture for AI agents that are…
Long-Running Agent
The Long-Running Agent Pattern defines the architecture for AI agents that execute…
Reflexive Agent
The Reflexive Agent Pattern defines an architecture in which an AI agent…
Single Agent Pattern
The Single Agent Pattern defines the canonical architecture for a single autonomous…
Stateful Agent Memory
The Stateful Agent Memory Pattern defines the multi-tier memory architecture that allows…
CodeAct Agent
The CodeAct Agent Pattern defines an architecture in which an agent's primary…
Conditional Routing
The Conditional Routing Pattern defines control flow structures for agentic workflows: if/else…
Context Compression
The Context Compression Pattern defines techniques for managing an agent's context window…
Dynamic Sub-agent Spawning
The Dynamic Sub-agent Spawning Pattern defines an architecture in which a parent…
Iterative Constraint Satisfaction
The Iterative Constraint Satisfaction Pattern defines an architecture in which an agent…
Mixture of Agents
The Mixture of Agents (MoA) Pattern defines an architecture in which multiple…
Parallel Fan-Out / Fan-In
The Parallel Fan-Out / Fan-In Pattern defines an architecture in which a…
Plan-and-Execute
The Plan-and-Execute Pattern defines a two-phase agentic architecture: a dedicated Planner decomposes…
ReAct Agent Loop
The ReAct (Reason + Act + Observe) Agent Loop Pattern defines an…
Router / Dispatcher
The Router / Dispatcher Pattern defines a central classifier that receives all…
Sequential LLM Chain
The Sequential LLM Chain Pattern defines a deterministic multi-step pipeline in which…
Streaming Progressive Output
The Streaming Progressive Output Pattern defines how agentic workflows deliver partial, incremental…
Tool Call Orchestration
The Tool Call Orchestration Pattern defines the execution mechanics for structured tool…
Workflow State Machine
The Workflow State Machine Pattern defines an explicit finite-state machine (FSM) that…
Workflow Tracing and Replay
The Workflow Tracing and Replay Pattern defines the instrumentation, storage, and replay…
APRA CPS230 AI Compliance
APRA Prudential Standard CPS230 Operational Risk Management (effective 1 July 2025) imposes…
EAAPL-CMP002 — APRA CPS234 AI Security Architecture
Australian Prudential Regulation Authority (APRA) Prudential Standard CPS234 Information Security applies in…
EAAPL-CMP007 — Data Residency for AI Architecture
AI systems introduce new vectors for cross-border data transfer that existing data…
EAAPL-CMP003 — EU AI Act Compliance Architecture
The EU Artificial Intelligence Act (Regulation (EU) 2024/1689) establishes legally binding obligations…
EAAPL-CMP008 — GDPR-Compliant AI Architecture
The General Data Protection Regulation (Regulation (EU) 2016/679) applies in full to…
EAAPL-CMP005 — ISO/IEC 42001 AI Management System Implementation
ISO/IEC 42001:2023 is the first international standard providing requirements for an Artificial…
EAAPL-CMP006 — NIST AI Risk Management Framework Implementation
The NIST AI Risk Management Framework 1.0 (NIST AI RMF, published January…
EAAPL-CMP004 — Privacy-Preserving AI Architecture
AI systems that ingest, process, or generate personal information must comply with…
AI Data Mesh Integration
Enterprises adopting Data Mesh architectures face a structural challenge when integrating AI:…
AI Training Data Governance
AI training data governance is the foundation of responsible AI.
Data Lineage for AI
Regulators, auditors, and risk committees increasingly demand that organisations explain not just…
Data Quality for AI
AI systems are uniquely sensitive to data quality failures in ways that…
Federated Learning Pattern
Federated learning enables multiple organisations or devices to collaboratively train a shared…
Privacy by Design for AI Data
Privacy-by-design mandates that privacy controls are embedded into AI data pipelines at…
Real-Time Feature Engineering
AI models that make real-time decisions — fraud detection, personalised recommendations, credit…
Synthetic Data Generation
Many of the most valuable enterprise AI use cases — fraud detection,…
AI Approval Workflow
The AI Approval Workflow establishes a multi-stage, evidence-gated approval process that every…
AI Audit Trail
The AI Audit Trail pattern establishes an immutable, tamper-evident log of every…
AI Ethics Review Board
The AI Ethics Review Board (AERB) is a standing organisational governance structure…
AI Incident Management
The AI Incident Management pattern implements a specialised incident management lifecycle for…
AI Model Register
The AI Model Register is the foundational governance artefact for any enterprise…
AI Policy Enforcement
The AI Policy Enforcement pattern implements runtime governance controls that ensure AI…
AI Regulatory Compliance Gateway
The AI Regulatory Compliance Gateway is a centralised, runtime compliance validation layer…
AI Risk Assessment Framework
The AI Risk Assessment Framework provides a systematic, repeatable methodology for evaluating…
Model Bias Detection
The Model Bias Detection pattern implements a continuous pipeline for detecting, measuring,…
Responsible AI Framework
The Responsible AI Framework operationalises the five core responsible AI principles—fairness, transparency,…
Active Learning Loop
The Active Learning Loop pattern establishes a structured, closed-loop process by which…
AI Confidence Threshold Routing
The AI Confidence Threshold Routing pattern dynamically routes inference requests to different…
Annotation and Feedback Loop
The Annotation and Feedback Loop pattern defines the end-to-end architecture for collecting…
Collaborative AI Decision
The Collaborative AI Decision pattern defines an architecture in which humans and…
Human Approval Gateway
High-stakes AI systems — those authorising transactions, making clinical recommendations, or influencing…
Human Escalation Pattern
The Human Escalation Pattern defines the architecture for routing AI-handled requests to…
Human Override Pattern
The Human Override Pattern provides a lightweight, universally applicable mechanism by which…
Hybrid Intelligence Pattern
The Hybrid Intelligence Pattern defines an architecture for systematically allocating each component…
EAAPL-INT007 — AI Circuit Breaker
EAAPL-INT003 — AI-Powered API Composition
AI-Powered API Composition uses large language models with function-calling capability to dynamically…
EAAPL-INT006 — AI Webhook Pattern
The AI Webhook Pattern delivers AI-generated notifications, inference results, and asynchronous AI…
EAAPL-INT005 — Batch AI Processing
Batch AI Processing applies AI inference to large volumes of data through…
EAAPL-INT008 — Bidirectional AI Sync
EAAPL-INT001 — Enterprise AI Service Bus
The Enterprise AI Service Bus pattern establishes an event-driven integration backbone that…
EAAPL-INT002 — Legacy System AI Augmentation
The Legacy System AI Augmentation pattern adds AI capabilities to existing enterprise…
EAAPL-INT004 — Real-Time AI Stream Processing
Real-Time AI Stream Processing applies AI inference directly to streaming data pipelines,…
EAAPL-KNW003: AI Knowledge Corpus Management
The AI Knowledge Corpus Management pattern defines the complete operational lifecycle for…
EAAPL-KNW006: Corpus Quality Assurance
Corpus Quality Assurance (CQA) is the automated pipeline that evaluates the fitness…
EAAPL-KNW001: Enterprise Knowledge Graph
An Enterprise Knowledge Graph (EKG) is a persistent, curated, machine-readable representation of…
EAAPL-KNW005: Knowledge Graph for Explainability
The Knowledge Graph for Explainability pattern uses a structured knowledge graph to…
EAAPL-KNW002: Semantic Data Layer
The Semantic Data Layer (SDL) is a governed translation layer that sits…
EAAPL-KNW004: Vector Database Management
Vector databases are the retrieval backbone of enterprise AI systems.
MCP Server Design
The MCP Server Design Pattern describes how to design, implement, and operate…
MCP Gateway
The MCP Gateway Pattern introduces a dedicated intermediary layer between MCP clients…
Multi-Server Orchestration
The MCP Multi-Server Orchestration Pattern describes how an AI model or agent…
MCP Authentication & Authorisation
The MCP Authentication & Authorisation Pattern defines the identity, authentication, and authorisation…
Stateful MCP Sessions
The Stateful MCP Sessions Pattern defines how to design, manage, and secure…
EAAPL-MDL003 — Canary Model Release
Canary model release is the controlled, incremental exposure of a new AI…
EAAPL-MDL006 — Fine-Tuning Pipeline
Fine-tuning is the process of adapting a pre-trained foundation model to an…
EAAPL-MDL008 — Model Access Governance
Model access governance controls who can invoke, export, copy, and administer AI…
EAAPL-MDL007 — Model Compression and Optimisation
Model compression reduces the computational footprint of an AI model — memory…
EAAPL-MDL004 — Model Rollback
Model rollback is the capability to revert production AI model serving to…
EAAPL-MDL001 — Model Versioning and Lineage Tracking
AI models are long-lived, continuously evolving artefacts.
EAAPL-MDL005 — Multi-Model Ensemble
Multi-model ensemble combines the outputs of two or more AI models to…
EAAPL-MDL002 — Shadow Model Deployment
Shadow model deployment allows an organisation to validate a new AI model…
EAAPL-MAG006 — Agent Handoff Protocol
EAAPL-MAG004 — Agent Swarm
EAAPL-MAG005 — Debate Agent
EAAPL-MAG003 — Human-in-the-Loop Agent
EAAPL-MAG001 — Multi-Agent Orchestration
EAAPL-MAG002 — Supervisor Agent
A/B Model Evaluation
Upgrading the LLM powering a production AI application carries risk that cannot…
EAAPL-OBS006 · AI Cost Observability
AI inference costs have a fundamentally different cost structure than traditional compute:…
EAAPL-OBS004 · AI Incident Management
AI system failures are qualitatively different from traditional software failures.
EAAPL-OBS008 · AI Performance Benchmarking
AI system quality degrades silently between benchmarking events.
EAAPL-OBS001 · AI Telemetry Architecture
AI systems present unique observability challenges that traditional APM tooling does not…
EAAPL-OBS007 · Distributed AI Tracing
AI pipelines are not single API calls.
EAAPL-OBS003 · Hallucination Detection
Large language models fabricate plausible-sounding content with confidence.
LLM Evaluation Pipeline
Deploying an LLM update to production without a structured evaluation gate is…
EAAPL-OBS005 · Model Drift Detection
AI models degrade silently.
Prompt Drift Detection
LLM providers update their underlying models without breaking API contracts.
EAAPL-OBS002 · Prompt Monitoring
Prompts sent to large language models in production are the primary control…
AI API Gateway
The AI API Gateway pattern establishes a purpose-built control plane that sits…
EAAPL-PLT010 — AI Developer Portal Architecture
Enterprise AI capabilities—LLM APIs, embedding services, vector stores, AI agent frameworks, fine-tuning…
AI Experiment Tracking
AI systems that cannot demonstrate systematic evaluation before production deployment are a…
Enterprise AI Platform
The Enterprise AI Platform pattern establishes a shared, governed infrastructure layer that…
Feature Store Integration
Feature stores solve a deceptively simple problem: when an ML model needs…
LLM Caching Layer
LLM inference is expensive and often redundant.
LLM Cost Control
LLM inference costs exhibit a dangerous property shared with no previous enterprise…
Model Routing
The Model Routing pattern establishes intelligent, policy-driven dispatch of AI inference requests…
Multi-Tenant AI Platform
When an AI platform serves multiple tenants—whether internal business units within an…
Prompt Version Control
Prompts are the primary programming interface for LLM-based systems, yet most organisations…
Agentic Retrieval-Augmented Generation
Agentic RAG places a reasoning AI agent in the orchestration loop of…
Contextual RAG with Metadata Filtering
Contextual RAG with Metadata Filtering extends the foundational RAG pattern with a…
Enterprise Retrieval-Augmented Generation
Retrieval-Augmented Generation (RAG) is the foundational architecture pattern that grounds Large Language…
Federated Retrieval-Augmented Generation
Federated RAG enables retrieval across multiple organisationally or geographically distributed knowledge bases…
Graph Retrieval-Augmented Generation
Graph RAG combines vector-based semantic retrieval with structured knowledge graph traversal to…
Hybrid Retrieval-Augmented Generation
Hybrid RAG combines dense (semantic) vector retrieval with sparse (keyword-based BM25) retrieval…
Multi-Source Retrieval-Augmented Generation
Enterprise knowledge is never stored in a single system.
Multimodal Retrieval-Augmented Generation
Multimodal RAG extends the retrieval-augmented generation paradigm to knowledge corpora that include…
Secure Retrieval-Augmented Generation
Secure RAG is the enterprise access-control overlay that must be applied to…
Streaming Retrieval-Augmented Generation
Streaming RAG extends the foundational RAG architecture to continuously ingest and index…
Extended Thinking Gate
The Extended Thinking Gate is an inference-time routing pattern that activates a…
Think Budget Allocation
Think Budget Allocation is the pattern of dynamically sizing the thinking-token budget…
Reasoning-then-Act
Reasoning-then-Act (R-then-A) is an agentic architecture pattern that separates the thinking phase…
Cost-Quality Router
The Cost-Quality Router is an LLM gateway pattern that dynamically selects from…
Multi-Step Verification
Multi-Step Verification is the pattern of using one or more independent reasoning…
Adversarial Input Defence
Adversarial Input Defence addresses a class of attacks specifically designed to manipulate…
AI Data Classification
AI Data Classification defines the automated detection, labelling, and enforcement architecture that…
AI Gateway
The AI Gateway pattern establishes a centralised, enterprise-grade control plane through which…
AI Output Filtering
AI Output Filtering is the post-generation inspection and transformation pipeline that evaluates…
LLM Input Sanitisation
LLM Input Sanitisation is a pre-processing pipeline that transforms raw application inputs…
Model Isolation
Model Isolation defines the architectural pattern for constraining the execution environment of…
Prompt Firewall
The Prompt Firewall is an inline defensive layer that inspects every user…
Secrets Management for AI
Secrets Management for AI addresses one of the most prevalent and consequential…
Secure Tool Invocation
Secure Tool Invocation defines the security architecture for AI agents that can…
Zero-Trust AI Pipeline
The Zero-Trust AI Pipeline applies the "never trust, always verify" security architecture…