Overview
OutLayer enables developers to run heavy computation off-chain while maintaining blockchain-grade security and on-chain payments. It’s not another L2—it’s an extension layer for NEAR that lets you break free from gas limits and execute complex logic (AI inference, API calls, ZK proofs, ML models) with cryptographic proof that the code ran correctly. Intel TDX hardware attestation on every action, user secrets protected via NEAR MPC key derivation, and built-in stablecoin monetization—all out of the box.
How It Works
When you call OutLayer, your request specifies a GitHub repo + commit hash or an immutable WASM binary. The system validates access conditions, derives decryption keys through the NEAR MPC Network, injects secrets as environment variables inside the TEE, executes your code in hardware-isolated memory, and returns results with a cryptographic attestation report. Anyone can verify exactly what code produced the output.
For NEAR smart contracts, OutLayer integrates via the native yield/resume mechanism—your contract pauses, computation runs off-chain in the TEE, and results return directly to L1. No bridge, no separate consensus, no L2 complexity.
Security Architecture
OutLayer combines two independent security layers:
Intel TDX (Trusted Execution Environment): Code runs inside CPU-encrypted memory that even the server operator cannot access. Every execution produces a hardware-signed attestation report—cryptographic proof that your code ran with your inputs in a tamper-evident environment. Secrets never touch the host OS.
NEAR MPC Network: 27 independent validators running threshold BLS12-381 cryptography (20-of-27 threshold). Key derivation is distributed—no single node ever sees the complete secret. This is the same infrastructure securing NEAR Intents and billions in cross-chain assets.
To compromise a user’s secret, an attacker must simultaneously break Intel TDX hardware isolation AND compromise 20+ independent MPC validator nodes operated by competing organizations. This dual-layer model makes OutLayer practically unbreakable.
Core Features
- Hardware Attestation: Every execution returns an Intel TDX attestation report proving code integrity
- Confidential Key Derivation: Derive encryption keys via NEAR MPC Network—no single point of failure
- TEE Secret Storage: Store API keys, private keys, and sensitive data in hardware-encrypted memory with access control via NEAR balance, NFT ownership, or whitelist
- Dual Integration Mode: Call from NEAR smart contracts (yield/resume) or any app via HTTPS API—same security, same attestation
- Built-in Monetization: Charge microtransactions per call using payment keys—no Stripe, no payment processor needed
- AI-Friendly Design: Standard WASI interface lets AI agents generate business logic in plain Rust while OutLayer handles TEE, MPC, and key management automatically
What You Can Build
- Verifiable AI Inference: Run OpenAI, Claude, or local models off-chain with attestation-backed results
- Attestation-Backed Oracles: Fetch real-time prices from any API, publish TEE-verified feeds
- Private DAO Voting: Encrypted votes counted inside TEE—nobody sees individual votes, not even operators
- Verifiable Randomness: Hardware-attested random numbers for games, lotteries, NFT minting
- Monetized APIs: Wrap external APIs, run custom logic, earn stablecoins per call
- Wallet-Based Apps: Private keys live in TEE—build signing services, MPC-backed auth with secrets persisting across upgrades
Why Not Just Use an L2?
L2s require separate consensus and bridging. Oracles only fetch data—they can’t run your code. Cloud functions ask you to “trust us” with no proof. OutLayer is different: results return directly to NEAR L1 with hardware attestation, you can execute any user code (not just data fetching), and monetization is built in. It’s the off-chain compute layer NEAR has been missing.
Updated: February 4, 2026
