# Integra's Three-tier Architecture Integra is structured as a three-tier stack that separates consensus, shared infrastructure, and end-user protocols. This design keeps the base chain simple and robust while allowing identity, compliance, and market applications to evolve independently.
Together, these tiers provide a vertically integrated environment where regulated real estate workflows can run end-to-end on a single, interoperable substrate. **Tier 1** #### Foundational Blockchain (Cosmos SDK + Ethermint EVM) Integra operates an EVM-compatible, Cosmos-based blockchain optimized for regulated real estate workflows. The network targets predictable block times and deterministic finality, making it suitable for primary issuance, secondary trading, and escrow events. Core network parameters – such as validator set size, block time, unbonding period, and slashing rules – are managed through on-chain governance and documented in the technical specification. The chain uses a delegated proof of stake (DPoS) model with a curated validator set composed of regulated infrastructure providers, trusted institutional validators, and other identifiable entities that meet strict operational and compliance criteria. Validators are subject to minimum self-stake requirements, slashing for double signing and downtime, and ongoing performance monitoring. This model balances decentralisation with the accountability expectations of institutional real estate partners. The execution layer is provided by Ethermint and offers full Ethereum bytecode compatibility. Existing Solidity contracts can be deployed without modification, and standard tooling such as MetaMask, Hardhat, Foundry, and Remix can be used out of the box. This lowers integration cost for RWA protocols and enterprise developers while preserving the sovereignty and customisability of a Cosmos blockchain. *** **Tier 2** #### Infrastructure Services Tier 2 hosts shared services that sit between the base chain and the application layer. These services provide identity, compliance, and data attestation and are invariably consumed by multiple applications. With the introduction of AI agents as first-class network participants, Tier 2 has been extended beyond identity, compliance, and data attestation to include a dedicated AI infrastructure layer. This layer provides the payment rails, inference attestation, settlement primitives, and security controls that AI agents require to operate reliably and accountably within regulated real estate workflows. All Tier 2 services are consumed by both human-facing and agent-facing applications through a unified interface. **a. Identity and Compliance Layer** Integra adopts W3C Decentralized Identifiers (DID) and Verifiable Credentials (VC) as the foundation for identity and access control. Participants – such as investors, issuers, brokers, custodians, and validators are represented by DIDs. Regulated entities (for example, KYC providers, registries, and regulated intermediaries) issue verifiable credentials that encode attributes such as KYC status, accreditation, jurisdiction, sanctions screening, or licence type. Compliance policies are expressed as machine-readable rules that reference these credentials and asset-level restrictions. Before a transaction is accepted on chain, the compliance engine validates that the sender, receiver, asset, and trade type satisfy applicable rules (for example, investor eligibility, transfer restrictions, lock-up periods, or travel rule requirements). These checks can be enforced at the smart contract level, as pre-execution hooks in the EVM, or via Cosmos modules that gate state transitions, depending on the use case. The identity layer is extended to cover AI agents as a distinct participant class. Each agent is assigned a DID bound to its ERC-8004 Agent Passport. Principal authorization is enforced via ERC-8118. Agent credentials encode behavioral scope permitted asset classes, maximum transaction value, and geographic restrictions evaluated in the same pre-execution compliance check as human credentials. See Agent Passport and ERC-8118 sections for full specification. **b. Data Attestation Services** Real estate requires high-quality off-chain data: title information, valuations, audits, covenants, and property management records. Integra provides a standard attestation framework where authorised attesters (such as registries, valuers, auditors, and notaries) sign structured attestations that are anchored on-chain. References to underlying documents are stored as content hashes or encrypted references rather than raw files, enabling integrity verification without exposing sensitive information. The attestation framework is extended to cover AI-generated outputs alongside human-attested data. When an AI agent produces a valuation, a due diligence summary, or a compliance assessment, the output is structured as a signed attestation anchored on-chain including the model version, input data hash, output hash, confidence interval, and a reference to any human reviewer sign-off where required by regulation. This means AI outputs are not stored as raw inference results but as verifiable, auditable attestations in the same framework as title records and valuation reports. Institutional participants can verify the provenance and methodology of any AI-generated data point before relying on it. **c. AI Payment Infrastructure** AI agents require a payment layer that is gasless, programmable, and auditable. Integra's Tier 2 payment infrastructure provides three primitives for agent-driven transactions: Fee Grants (x/feegrant) Cosmos fee grant module enables agent operators or dApp sponsors to cover transaction fees on behalf of agents and end users. Agents operate gasless by default, the sponsoring principal's IRL balance is debited, not the agent's wallet. This eliminates the "chicken and egg" problem of agents needing tokens to act before they have earned any. Programmable Commission Routing Commission splits between Seller Agent, Buyer Agent, Affiliate Agent, and the protocol, are encoded in the listing at creation time and executed atomically at settlement. No post-hoc payment logic. Every fee destination is on-chain and verifiable before a deal closes. Stablecoin Settlement Rail All agent-to-agent and agent-to-human settlements are denominated in Integra's native stablecoin (USDC bridged via IBC as primary until native issuance is live). Agents never hold volatile assets for settlement only IRL for staking and the settlement stablecoin for deal execution, keeping balance sheet risk manageable for both agents and their principals. **d. AI Settlement Layer** Deterministic finality from CometBFT is the foundation, but agent-specific settlement primitives sit above it: **Atomic deal execution:** the full deal lifecycle compliance check, asset transfer, stablecoin payment, commission split, Asset Passport update executes in a single atomic transaction. There is no intermediate state where payment has occurred but title has not transferred, or vice versa. Agents cannot be partially settled. **Escrow module:** for deals requiring human review or regulatory sign-off before final settlement, the escrow module holds funds and asset control in a time-locked on-chain vault. Either party can trigger release or initiate dispute resolution within the defined window. The escrow state is fully visible to both agents and their principals. **Proof-of-payment anchoring:** every stablecoin settlement produces an on-chain proof-of-payment record linked to the corresponding Asset Passport entry. This satisfies AML transaction monitoring requirements and provides the audit trail that institutional compliance teams require for cross-border deals. **Yield distribution:** for income-generating assets, the settlement layer includes an automated yield distribution module. Rental income or dividend payments are distributed to token holders on a defined schedule, without requiring any manual action from asset managers or agents. Distribution events are recorded in the Asset Passport transaction history. **e. Agent Scope Enforcement** Integra's agent authorization model defines what an agent is permitted to do -- asset classes, geographic markets, maximum transaction sizes, permitted counterparty types. When an agent submits a transaction, the protocol validates it against the agent's authorized scope before execution. Transactions outside scope are rejected at the contract level, regardless of what triggered them. This provides a hard boundary against agent misbehavior: even if an agent's off-chain logic is compromised, manipulated, or produces unexpected outputs, the on-chain authorization layer enforces the principal's constraints. The agent cannot spend funds, trade assets, or interact with contracts beyond what its principal explicitly authorized. **AI Output Validation** For regulated parameters (compliance status, jurisdiction checks, transaction limits), deterministic on-chain checks override AI-generated outputs. Agent reasoning and language generation execute off-chain; only structured, schema-validated outputs are submitted for on-chain execution. **f. AI Inference Infrastructure** AI inference does not occur on-chain storing raw LLM reasoning on a blockchain is astronomically expensive and architecturally unnecessary. Instead, Integra uses an off-chain inference with on-chain attestation model: **Off-chain inference:** agent reasoning, valuation models, compliance normalization, and natural language negotiation all execute off-chain on operator-managed or cloud-hosted inference infrastructure. Operators are free to use any LLM provider compatible with the Integra Agent SDK. **On-chain output attestation:** the final output of any agent decision that triggers a state transition is hashed and signed as an on-chain attestation before execution. This creates a verifiable record of what the agent decided and on what basis, without storing the full inference trace on-chain. **Model versioning:** agent behavior specifications including model version, system prompt hash, and tool configuration are stored on-chain and linked to the Agent Passport. Changes to an agent's model configuration require a new on-chain registration event, creating an auditable version history. **Cost model at scale:** at 10,000 simultaneous active agents averaging 100 decisions per day, inference costs depend on operator model choice. Integra's SDK is designed to work with efficient inference endpoints (Llama 3, Mistral, or GPT-4o-mini class models) where cost per decision is $0.0001–$0.001, giving a total infrastructure cost of $100-$1,000/day at this scale well below the intermediary fees being replaced. Agent operators are responsible for their own inference infrastructure. Integra does not host or operate agent execution logic -- the protocol provides identity, authorization, and settlement infrastructure that agents plug into. *** **Tier 3** #### Core Protocol Applications Tier 3 contains Integra’s native applications. These are built on top of the foundational chain and consume Tier 2 services for identity, compliance, and data. The initial core modules are: [Asset Passport](https://whitepaper.integralayer.com/core-native-dapps/asset-passport) [Agent Passport](/core-native-dapps/agent-passport.md) [iRWA Token Bridge](/core-native-dapps/irwa-token-bridge-ai-token-bridge.md) [Global Orderbook](https://whitepaper.integralayer.com/core-native-dapps/global-orderbook) [Native Stablecoin](https://whitepaper.integralayer.com/core-native-dapps/native-stablecoin) #### Summary View of the Three Tiers * Foundational L1 Blockchain * Cosmos SDK with Ethermint EVM execution * Curated, regulated validator set with DPoS and slashing * Deterministic finality and full Ethereum tooling compatibility * Middle Layer – Cross-Application Infrastructure * Identity and compliance layer based on DIDs and verifiable credentials * Data attestation framework for title, valuations, audits, and other off-chain records * Core Native DApps * Asset Passport registry for standardised, verifiable asset profiles * Global Orderbook for cross-venue liquidity, routing, and commission tracking * Native stablecoin for compliant on-chain settlement and pricing This three-tier architecture grounds Integra’s positioning as a real estate-specific Layer-1: the base chain remains general-purpose and EVM compatible, while the infrastructure and application tiers encode the compliance, data, and settlement primitives required for institutional-scale real estate markets. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://whitepaper.integralayer.com/integra-the-real-estate-layer-1/integras-three-tier-architecture.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.