Solana Foundation Grant Application - Fully Onchain Perps CLOB
Status: Ready for submission Company: Artifi Labs Date: July 2026 Total ask: $65,000 (Milestone-based grant)
1. Company / Project Name
Company: Artifi Labs Project: Fully Onchain Perps CLOB
2. Website URL
[TBD - project website to be created during M1]
Supporting materials during application review:
- Build plan:
docs/build-plan.md - Milestones:
docs/milestones.md - Risks & security:
docs/risks-and-security.md
3. Country
Portugal
4-6. Applicant Legal Name / Email
| Field | Value |
|---|---|
| First Name | Kylix Alexander |
| Last Name | Da Silva Alves Afonso |
| kylix@artifi.finance |
7. Solana On-Chain Accounts
N/A
8. Funding Amount
$65,000 (Milestone-based grant)
9. Funding Category
DeFi
10. Project / Idea Description
What We Are Building
A fully onchain perpetual futures protocol on Solana, anchored by a central limit order book (CLOB) with price-time priority matching. Every order placement, cancellation, match, and settlement executes as a Solana transaction - no offchain sequencer or matching engine.
Why This Matters
Most perps volume today flows through either centralized exchanges or hybrid architectures that rely on offchain sequencers. Solana is the first chain with the throughput and latency to support a fully onchain CLOB for derivatives. We are building the reference implementation.
Key Design Decisions
| Decision | Choice | Rationale |
|---|---|---|
| Architecture | Onchain CLOB, slab-based, price-time priority | Maximum transparency; price discovery via two-sided flow |
| Matching model | Fully onchain | No offchain sequencer or matching engine |
| Initial market | SOL-PERP | Highest liquidity demand on Solana |
| Margin mode | Isolated margin only | Prevent socialized losses; simpler risk model |
| Collateral | USDC only | Most widely used Solana stablecoin |
| Licensing | BSL 1.1 (programs + SDK) | Source-available - prevents commercial closed-source forks, maximizes public good |
| Oracle | Pyth (EMA + deviation guards) | Battle-tested on Solana; confidence-interval support |
Innovation
- Isolated-margin CLOB with onchain insurance fund + ADL fallback - fully onchain perps CLOB on Solana with this specific architecture. Most existing Solana perps, such as Jupiter Perps, GMTrade, Flash Trade, etc, use AMM/vAMM models or are hybrid (some require offchain components). A slab-based CLOB with price-time priority offers superior price discovery for institutional flow.
- Explicit ADL backstop - if the insurance fund is exhausted, auto-deleveraging is triggered programmatically, preventing insolvency without governance intervention.
Why Solana?
Solana is the only chain today with the throughput, latency, and ecosystem maturity to support a fully onchain perps CLOB:
- Throughput: Solana's SVM can process thousands of orders per second, matching the cadence of a real-time order book. No other general-purpose chain comes close - Ethereum L1 cannot sustain the state writes of an active CLOB, and L2s add sequencer latency that undermines price-time priority.
- Latency: Sub-second block times mean orders are matched within one slot. This is enough for retail and most institutional strategies. Combined with priority fees, traders can bid for execution speed when it matters.
- Ecosystem: Pyth oracles, SPL tokens, Geyser plugins, and a mature Anchor toolchain already exist. We are building on proven infrastructure, not inventing new primitives.
- Culture: Solana's ethos of "ship in the open" aligns with our open-source commitment. The Foundation's criteria (fully onchain execution, real price discovery, Solana-first) describe exactly what we are building.
Alignment with Foundation Criteria
| Foundation criterion | How this project meets it |
|---|---|
| 1. Fully onchain execution | Every order, match, and settlement is a Solana tx. No offchain sequencer or matching engine. |
| 2. Real price discovery | Slab-based CLOB with price-time priority. Price comes from two-sided flow, not a pool formula. |
| 3. Solana-first | Built exclusively for SVM using Pyth, SPL, Geyser, and priority fees. 10% of protocol revenue hardcoded to Solana public goods. |
| 4. Innovation | Isolated-margin CLOB with onchain insurance fund + ADL on Solana. Reference implementation for fully onchain perps. |
| 5. Experienced team | Solana-native engineers with mainnet deployments, spot order-book DEX experience, and liquidation bot expertise. |
| 6. Open source | Core protocol (programs + SDK) BSL 1.1 from day one. Source-available, publicly verifiable. |
11. Funding Amount & Milestones
$65,000 (Milestone-based grant)
Milestone Overview
| # | Focus | Budget |
|---|---|---|
| M1 | Research, architecture & visual prototype | $5,000 |
| M2 | Core on-chain programs & devnet integration | $15,000 |
| M3 | Complete on-chain programs | $12,000 |
| M4 | Off-chain infrastructure (indexer, keeper) | $10,000 |
| M5 | Full trading interface & monitoring | $8,000 |
| M6 | Security & mainnet | $10,000 |
| M7 | Developer SDK & ecosystem | $5,000 |
| Total | $65,000 |
See Milestones for the full breakdown.
M1 - Research, Architecture & Visual Prototype ($5,000)
Architecture specification, threat model, and working visual prototype with mock data demonstrating wallet connect, trade page, and order form.
M2 - Core On-chain Programs & Devnet Integration ($15,000)
Perps and Vault programs deployed to Solana devnet with complete order lifecycle. Video walkthrough: deposit, place order, fill, close position, withdraw.
M3 - Complete On-chain Programs ($12,000)
Funding rate settlement, liquidations, insurance fund, ADL fallback, full admin controls. Published Anchor IDL and comprehensive test suite.
M4 - Off-chain Infrastructure ($10,000)
Indexer for on-chain events and Keeper Bot (liquidation + market-making). API server for unsigned transaction building.
M5 - Full Trading Interface & Monitoring ($8,000)
Complete web app (trade, portfolio, vaults, leaderboard) with real-time updates. Monitoring dashboards and alerting. Public devnet demo. Adoption target: 500 wallets or 5,000 trades in first week.
M6 - Security & Mainnet ($10,000)
Multi-sig governance, timelock mechanism, SEAL Framework certification, and mainnet deployment. Independent security audit funded separately by the Solana Foundation (see Section 20).
M7 - Developer SDK & Ecosystem ($5,000)
TypeScript SDK published to npm with provenance. Integration examples and documentation on protocol docs site.
12. Open Source Status
Partially. The core protocol is open source under a single license.
- License: BSL 1.1 (source-available - prevents commercial closed-source forks during the license period)
- Open source: Programs and TypeScript SDK
- Public from M1 with CI and buildable from a clean clone
- Open-source release is a formal M5 deliverable; source remains public and verifiable
13. Relevant Metrics
The project is pre-launch - there are no user-facing metrics to report at this stage. The grant itself funds the creation of the protocol, after which the following metrics will be tracked and reported:
14. Funding Status
Planning to raise funds in the next year.
This is the first funding round for this project. No prior grants, venture capital, or revenue. The Solana Foundation grant would fund the full MVP build, mainnet launch, and first quarter of operations.
15. Competition
The Solana perps ecosystem already has a few important players. Jupiter Perps and Adrena, which utilize unified, trader-vs-LP pool models to offer deep, single-pool liquidity at the cost of asset variety. Flash Trade and GMTrade iterate on this with multi-pool and isolated-pool structures (such as GMTrade’s Global Liquidity Vault) to let LPs choose their risk exposure while scaling asset catalogs to include commodities, equities, and foreign exchange. Outside of the predmoninatly AMM-style protocols, there also exists order book based protocols like Phoenix, which delivers a fully onchain CLOB, and Pacifica with its hybrid approach of offchain matching and onchain settlement.
Outside Solana, dYdX runs on its own purpose-built chain, using validator-maintained in-memory order books for short-term orders with onchain settlement. Hyperliquid takes a different approach with its own high-performance L1, where HyperCore supports fully onchain perps and spot order books. These platforms show that there is strong demand for CLOB-based perps, but they sit outside the Solana ecosystem and therefore do not fully benefit from Solana-native composability.
Our key differentiators:
- Fully onchain execution - Unlike Pacifica, Jupiter, Adrena, GMTrade and dYdX, every order, match, and settlement occurs directly on Solana. No sequencer, no offchain engine, no trust assumptions.
- True CLOB price discovery - Jupiter, Flash Trade, Adrena, and GMTrade rely heavily on external oracles (Pyth, Chainlink, Chaos Labs) to determine prices. Our slab-based CLOB with price-time priority produces genuine two-sided price discovery. This matters for institutional flow that demands visible, time-priority book depth.
- Solana-native, not hybrid - Built exclusively for Solana SVM from day one. No bridges, no EVM compatibility, no external sequencer. Full use of versioned transactions, priority fees, Pyth oracles, and Geyser.
We respectfully acknowledge that pool-based giants like GMTrade or Jupiter and high-speed hybrids like Pacifica have achieved meaningful product-market fit. However, our bet is that the market will shift toward fully onchain CLOBs as traders demand verifiable execution - the same shift that moved spot volume from AMMs to onchain order books on Solana.
16. How Is This Project a Public Good?
This project generates public goods for the Solana ecosystem in four distinct ways:
1. Open-source core protocol. The on-chain programs (clob, vault, risk, oracle-adapter) and the TypeScript SDK are published under BSL 1.1. Any developer can read, audit, and build on top of our matching engine, vault, or risk modules. The core financial infrastructure - the parts that custody user funds and execute matches - is fully transparent and auditable.
2. Public good liquidity. The M6 phase includes liquidity mining incentives and MM partnerships specifically designed to bootstrap the SOL-PERP market. This liquidity is a public resource - any Solana user can trade against it. The reference MM bot is itself open-source, so anyone can run a market-making strategy.
3. Community-first launch. There is no private sale, no token allocation, and no VC round tied to this first launch. After mainnet launch, the protocol fee treasury (accumulated from taker fees) is designed to eventually fund further public goods - audits, additional markets, and open-source tooling - via a transparent, onchain-governed process.
17. Why You?
(100+ words required)
Artifi Labs is a 6-person team of experienced DeFi architects across different blockchains. While this is our first project in the Solana ecosystem, we bring deep experience from building some of the biggest Cardano protocols, including complex lending platforms and data indexing infrastructure.
Team: Kylix (CEO/engineer) - product/BD/strategy; Sal (CTO) - protocol architecture lead; Nuno, Nogueira, Gil - protocol engineers; Tim - designer.
Proven DeFi track record: We have a history of shipping high-integrity financial primitives in production. Our engineering team has designed and implemented risk engines, isolated-margin modules, vault systems, and protocol infrastructure that have collectively managed significant TVL on Cardano. Artifi Labs also developed Open Djed in the context of Project Catalyst Fund14, giving us hands-on experience with stablecoin mechanics, protocol integrations, and open-source DeFi infrastructure. We understand the exact adversarial scenarios, such as oracle manipulation, toxic liquidation flow, and insolvency risks, that define the success of a perps protocol.
Why the move to Solana? After evaluating where a fully onchain perps CLOB could actually work at scale, we came to a conclusion: Ethereum L1 can't handle the state writes. L2s introduce sequencer latency that breaks price-time priority. Solana is the only network where every order, match, and settlement can execute onchain within a single slot - with the throughput to match a real-time order book. The sub-second finality, Pyth oracles, and Anchor toolchain already exist. We're not building on Solana because it's popular but because the architecture we want to build - isolated-margin CLOB with onchain insurance fund - is only viable on Solana.
Transparency: We are transparent that our team's strongest capabilities are technical - protocol architecture, risk engines, and on-chain systems. We welcome partnership with Solana Foundation advisors and partners who can help shape the business model, go-to-market strategy, and institutional partnerships. We are fully open to working with Foundation-affiliated teams who have deeper DeFi domain expertise, and we would appreciate introductions to audit firms for security reviews.
18. Revenue Model & Solana Alignment
Addressing the Foundation criterion: "Application revenue should be structurally routed back to the chain, preferably set at protocol level from launch."
Grant Type: We are seeking a Milestone-Based Grant.
Protocol fees:
- Taker fee: 5 bps
- Maker rebate: 1 bps (net protocol revenue: 4 bps on taker flow)
- Liquidation penalty: 5% of position value (flows to insurance fund)
- Funding rate: Variable, paid long-to-short or short-to-long (zero-sum among traders)
Routing to Solana:
- 10% of net protocol revenue (retained protocol fees after maker rebates/payouts) will be sent to an onchain Solana Public Goods Treasury or SPL Governance-managed fund at protocol level from launch. This is hardcoded in the
riskprogram - no future governance vote required. - Fee schedule is governed by a 3-of-5 multi-sig, with a planned transition to SPL Governance once the protocol is stabilized.
This ensures that as the protocol scales, a portion of its revenue structurally benefits the broader Solana ecosystem - not just the protocol's treasury.
19. Audit & Security Plan
Security is the primary risk for any perps protocol. Our approach is staged across milestones:
| Phase | Security Activity |
|---|---|
| M1 | Threat model documented (top 10 attack surfaces) |
| M2 | Property-based tests for slab matching; Bankrun adversarial integration suite |
| M3 | Chaos tests; comprehensive test suite covering liquidation flow and oracle rejection |
| M6 | Independent security audit; SEAL Framework Checklist certification; Immunefi bug bounty; operational runbook; incident response plan |
Full professional audit is completed during M6 before mainnet deployment. Security is a prerequisite for launching any protocol that handles user funds. We request the Solana Foundation's support in covering the cost of the independent security audit (see Section 20).
20. Support Beyond Funding
We welcome the Solana Foundation's assistance with:
- Security audit funding - covering the cost of an independent security audit for on-chain programs
- Legal entity setup - establishing the legal entities that will control the protocol and any future token
- Audit introductions - connections to reputable security audit firms for reduced-cost.
- Technical mentorship - guidance from Foundation partners with deep DeFi and derivatives expertise