← 返回
AI智能
中文
Aave
Assist with Aave lending, borrowing, liquidations, and risk management across chains.
协助进行Aave跨链借贷、清算及风险管理
ivangdavila
AI智能
clawhub
v1.0.0 1 版本 100000 Key: 无需
#latest
概述
Core Concepts
- Supply assets to earn interest — deposit, receive aTokens representing position
- Borrow against collateral — must supply first, then borrow up to limit
- aTokens accrue interest — balance grows over time automatically
- Health Factor determines liquidation risk — below 1.0 = liquidation
- Variable and stable rates available — stable costs more but predictable
Health Factor (Critical)
- Health Factor = (Collateral × Liquidation Threshold) / Borrowed
- Above 1.0 is safe — higher is safer
- At 1.0, liquidation begins — partial position closed
- Monitor actively during volatility — prices move, health factor changes
- Add collateral or repay debt to improve — before liquidation happens
Supplying (Lending)
- Deposit supported assets — ETH, stablecoins, various tokens
- Receive aTokens 1:1 — aETH, aUSDC, etc.
- Interest accrues in real-time — aToken balance grows
- Can withdraw anytime if liquidity available — high utilization may block withdrawals
- Enable as collateral to borrow against — optional per asset
Borrowing
- Must have collateral supplied first — can't borrow without
- Borrow up to LTV (Loan-to-Value) ratio — varies by asset, usually 70-85%
- Interest accrues on borrowed amount — must repay more than borrowed
- Variable rate changes with market — stable rate fixed but higher
- Debt tokens represent borrowing — not transferable
Liquidations
- Triggered when Health Factor < 1 — automated, permissionless
- Liquidators repay portion of debt — receive collateral + bonus
- Liquidation penalty 5-10% — you lose this bonus amount
- Up to 50% of debt liquidated at once — may need multiple liquidations
- Prevention: monitor and manage HF actively
Multi-Chain Deployment
- Aave V3 on Ethereum, Polygon, Arbitrum, Optimism, Avalanche, more
- Same interface, different markets — assets and rates differ
- Bridged assets may differ — USDC vs USDC.e
- Portals enable cross-chain — supply on one chain, borrow on another
E-Mode (Efficiency Mode)
- Higher LTV for correlated assets — stablecoins to stablecoins, ETH to stETH
- Up to 97% LTV in E-Mode — vs ~80% normally
- Only borrow assets in same E-Mode category — restricted but efficient
- Higher liquidation risk — narrow margin, monitor closely
GHO Stablecoin
- Aave's native stablecoin — minted by borrowing
- Backed by Aave collateral — overcollateralized
- Interest paid to Aave DAO — different from regular borrowing
- stkAAVE holders get discount — reduced borrow rate
AAVE Token
- Governance token — vote on proposals
- Staking in Safety Module — earn rewards, risk of slashing in shortfall
- stkAAVE for staking — represents staked position
- 10-day cooldown to unstake — plus 2-day unstake window
Risk Management
- Don't max out borrowing — leave buffer for price movements
- Diversify collateral — single asset concentration increases risk
- Use stablecoins for lower volatility — stable collateral = stable HF
- Set alerts for Health Factor — services like DefiSaver
- Consider automation — automatic deleveraging tools
Common Mistakes
- Borrowing at max LTV — immediate liquidation risk
- Ignoring variable rate changes — rates can spike quickly
- Not monitoring during volatility — HF changes fast with price
- Supplying without enabling collateral — can't borrow if not enabled
- Forgetting about interest — debt grows over time
Gas Considerations
- Approvals needed for each new asset — first-time gas cost
- Supply and borrow are separate transactions — plan gas for both
- L2 deployments much cheaper — Arbitrum, Optimism save significantly
- Batch operations where possible — some aggregators help
Integrations
- DefiSaver for automation — auto-repay, auto-leverage
- Instadapp for advanced management — DeFi dashboard
- 1inch, Paraswap for swaps — swap and supply in one transaction
- Flash loans for advanced users — borrow without collateral, repay in same tx
版本历史
共 1 个版本
-
v1.0.0
当前
2026-03-29 02:33 安全 安全
安全检测
腾讯云安全 (Sanbu)
安全,无风险
查看报告
🔗 相关推荐
ai-intelligence
ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,349
📥 317,697
ai-intelligence
oswalpalash
类型化知识图谱,用于结构化智能体记忆与可组合技能。支持创建/查询实体(人员、项目、任务、事件、文档)及关联...
★ 709
📥 243,527
ai-intelligence
pskoett
捕获经验教训、错误和纠正,以实现持续改进。使用时机:(1)命令或操作意外失败;(2)用户纠正……
★ 4,055
📥 795,905