This proposal implements a modular L2 support layer within Atlas PAB. The deliverables include:

Hydra Support: Join, create, or monitor Hydra heads from within Atlas workflows.

Submit transactions into Hydra heads from dApp endpoints.

Synchronize state transitions and exit strategies back to Cardano L1.

Provide test cases for L1↔Hydra hybrid contract flows.

Midnight & Midgard Integration: Define Atlas interfaces for submitting confidential payloads to Midnight or Midgard.

Enable fetching proof results or outcomes from private L2s for verification in L1.

Build example “hybrid private” dApp endpoints (e.g., voting, encrypted asset flow).

ZkFold Integration: Interface Atlas with ZkFold’s proving and aggregation layer.

Enable off-chain transaction batching or logic execution with zk-proof submission to L1.

Provide zk circuit calling templates in Atlas emulator tests.

Each module will be developed as a plug-in under Atlas.L2 namespace, maintaining modularity and future extensibility. The proposal also includes example apps, documentation, and public education (via GitHub and docs site).

Cardano is expanding its scalability layer through Hydra, Midnight, ZkFold, and Midgard – each offering unique benefits around privacy, speed, or parallelization. However, no unified backend framework exists that enables developers to easily integrate these Layer 2 (L2) solutions into their smart contracts or off-chain logic.

Currently, teams building dApps that want to support Hydra heads, confidential computation (Midnight), or ZK-powered offloading (ZkFold) must build custom integrations from scratch. Atlas PAB – as the leading Haskell-native backend for Cardano – is ideally positioned to serve as a bridge between Cardano L1 and emerging L2s.

Without such support, dApp developers face high integration complexity, reduced composability, and slower adoption of L2s.

This proposal delivers:

Seamless Hydra integration in Atlas, enabling dApps to spin up, join, and transact within Hydra heads from off-chain logic.

Foundational integration with Midgard, Midnight, and ZkFold L2s for submitting payloads, querying outcomes, and synchronizing state with Cardano L1.

Shared L2 interface modules in Atlas that support plug-in logic per L2 solution.

Public documentation, emulator test cases, and example dApps demonstrating hybrid L1+L2 architecture.

The result is a next-generation backend for the Cardano ecosystem, enabling builders to leverage fast, private, or ZK-compressed execution environments — without leaving the Atlas framework.

• TX submission
• Emulator tests Run dApp logic in Hydra head via Atlas; submit result to L1
• Result polling
• Proof decode API Submit confidential message; verify response in Atlas backend
• ZK circuit call samples
• Aggregated submission Atlas testnet demo submitting a ZK-compressed batch TX to L1
• 3 hybrid dApp samples Published in Atlas docs; referenced in GitHub README

Hydra L2 Integration: 140 hrs = $30,000 Midnight / Midgard Integration: 120 hrs = $25,000 ZkFold Support: 90 hrs = $20,000 Developer Docs, Testing, dApp Examples: 40 hrs = $7,000 Audit / Project Management (8%) — $8,000 Total — $90,000

duration 7/8 months Hydra L2 Integration: 140 hrs Midnight / Midgard Integration: 120 hrs ZkFold Support: 90 hrs Developer Docs, Testing, dApp Examples: 40 hrs

The Genius Yield team are the original creators and maintainers of Atlas PAB, and are recognized contributors to open-source infrastructure in the Cardano ecosystem. Our work has helped shape core backend tooling used by developers across DeFi, identity, and emerging governance applications. We established a strong working relationships with IOG, ZkFold, and Midnight engineering teams, ensuring alignment with upcoming Layer 2 protocols, privacy infrastructure, and roadmap-driven innovation.

This L2 support will become part of the official Atlas repository https://github.com/geniusyield/atlas and will be maintained by the Genius Yield engineering team as part of our long-term commitment to the Cardano developer ecosystem. All modules will follow Atlas’s versioning and release schedule and evolve alongside Cardano L2 tooling.

Hydra is actively promoted by IOG as a key L2 scaling solution. Community testnets are live and growing, but tooling for smart contract developers is still early-stage.

Midnight is an anticipated privacy-preserving L2 that developers want to integrate into voting, finance, and identity dApps.

ZkFold, an ecosystem-led zero-knowledge compute and compression solution, has begun publishing tools and dev docs for programmable L2 design.

Atlas PAB is already used in major Cardano dApps (Genius Yield, World Mobile, Clarity...) and has strong community backing as a developer-friendly backend. This proposal expands its capabilities to meet the ecosystem’s L2 trajectory.

Active L2 documentation for: Hydra (IOG), ZkFold (community), Midnight/Midgard (Input Output)

L2 endpoints and SDKs as they stabilize

Continued Cardano node and Plutus version compatibility