@
Your DeFi Positions Defend Themselves
Lighthouse continuously evaluates position health against configurable thresholds. When danger appears, the protocol executes corrective actions — repay, exit, rebalance — without the user signing anything.
Pitch of the Core Idea
DeFi positions require constant monitoring. A lending position can get liquidated while you sleep. An LP position can bleed impermanent loss during a volatility spike. Most users don’t have the tools, time, or expertise to manage this actively. The onchain risk curator market reached $7B AUM in early 2026, with Steakhouse ($1.53B), Sentora ($1.34B), and Gauntlet ($1.29B) controlling 70%. Yet these serve institutions, not retail.
Tributary v1 enables automated risk management: Lighthouse continuously evaluates position health, and when danger appears, the protocol executes corrective actions. Users approve protective strategies at setup; the protocol executes within those bounds. Non-experts get institutional-grade protection.
Core Mechanics
- Risk management service configures ComposablePolicy instances encoding protective actions
- Pre-validation (Lighthouse) — Evaluates position health against thresholds:
- LTV ratio below safety margin?
- Impermanent loss exceeds max tolerance?
- Collateral value dropped below minimum?
- Oracle price moved beyond volatility band?
- Pull (Token Delegation) — Claims tokens for corrective action (USDC for repayment, LP token for exit)
- Forward (CPI) — Executes corrective action: repay loan, withdraw LP, swap to stablecoin
WHEN (condition: LTV > 75% danger threshold)
→ VALIDATE (Lighthouse: assert MarginFi LTV for user position > 0.75)
→ PULL (repayment amount USDC)
→ FORWARD (repay loan on MarginFi → reduce LTV to safe level)
Psychological Hook and Addictiveness
“I sleep while my positions defend themselves.” The anxiety reduction is enormous. No more 3am wake-up calls from liquidation alerts. No more checking positions every hour during volatility. The shield runs continuously, silently, effectively.
Risk dashboard: Real-time view of all protected positions, threshold status, corrective actions taken. Protection history: Log of every prevented liquidation, every IL mitigation — tangible proof of value. Confidence to deploy more capital: With protection in place, users allocate more to DeFi, earning higher yields.
Brief Market Research
| Metric | Data |
|---|---|
| Onchain Risk Curator AUM | $7B (2026) |
| Steakhouse AUM | $1.53B |
| Sentora AUM | $1.34B |
| Gauntlet AUM | $1.29B |
| DeFi TVL | $50B+ |
| Lending Protocol TVL | $20B+ |
Key Competitors:
- Gauntlet: Risk simulation for protocols, $1.29B AUM, institutional focus
- Steakhouse: RWA risk curation, $1.53B AUM, Coinbase backend
- Sentora: AI-driven risk models, $1.34B AUM, Kraken backend
- DeFi Saver: Single-protocol automation for Aave/Compound, but no Lighthouse validation
- Antidote: Margin call protection on Uniswap v4, but EVM-only
None provide non-custodial, Lighthouse-validated risk management across Solana DeFi.
Business Model
- Protection fee: 0.5-2% of protected capital annually
- Execution fee: 0.1-0.5% on each corrective action
- Premium thresholds: Advanced risk parameters for sophisticated users ($20-100/mo)
- Institutional tier: $500-2,000/mo for portfolio-wide risk management
- Insurance pool: Users contribute to mutual insurance pool for extreme events
Summary of Technical Specifications
Architecture
- Position monitor (reads lending/LP protocol state)
- Lighthouse validation layer (evaluates risk thresholds)
- Corrective action engine (repay, exit, rebalance)
- Token delegation system (pull for corrective actions)
- Forward CPI routing (execute corrective actions)
- Risk dashboard (position health, threshold status, history)
How This Hooks Into Tributary
- PayAsYouGo: Pulls tokens for corrective actions within delegated cap
- Lighthouse: Validates risk thresholds against position state
- Forward: CPI to lending/LP/DEX protocols for corrective actions
- ComposablePolicy: Defines risk thresholds, corrective actions, delegation caps
Recommended Tech Stack
- Solana + Anchor
- Tributary SDK (PayAsYouGo + Lighthouse + Forward)
- MarginFi, Kamino, Meteora SDK for position reading
- Pyth/Switchboard for price feeds
- Keeper service for threshold monitoring
- React dashboard for risk visualization
MVP Scope
- Single-protocol risk management (MarginFi lending)
- Auto-repay on LTV threshold breach
- Basic risk dashboard
- Simple keeper service
- Buildable in 2-3 days with Tributary v1 SDK + MarginFi
Non-Technological Requirements
- Keeper service reliability: Off-chain keeper triggers execution. Must be fast, reliable, decentralized
- Protocol integration: Need deep integration with lending/LP protocols for position state reading
- Threshold calibration: Over-protection triggers too early; under-protection misses danger. Key UX challenge
- User education: Users must understand protection is not guarantee — extreme events may still cause losses
- Insurance considerations: May need mutual insurance pool for black swan events
Potential Risks
- Validation lag: Keeper service triggers execution off-chain. If slow, position may liquidate between threshold-cross and execution
- Cascade risk: During market-wide crashes, many positions cross thresholds simultaneously. Gas spikes may delay protective actions
- Over-protection: Aggressive thresholds trigger protective actions too early — exiting positions that would have recovered
- Protocol-specific account layouts: Each lending/LP protocol has different account structures. Lighthouse assertions must be protocol-specific