Skip to main content
Sign in
Aave V33 decisions on this page

Audit log

Every state-changing event for Aave V3: moderation decisions on community submissions, plus corrections and updates from the news pipeline. URL-based decisions carry three independent witnesses — the original source, an Internet Archive snapshot taken at submission time, and a Solana memo signed by our publicly-disclosed publisher key.

  1. #1publishby system:backfill
    2026-05-29 16:00:01Z
    Score: ?? (no score change)
    anchoranchored
    chain
    mainnet-betaslot 422,970,397
    sig
    53ojCUZotuii…E6bboAZzexplorer ↗
    hash
    zyhqDXVECDiJ…vHRTa6nTsha256 → base58
    verifying row…full verify ↗
    canonical bytes (8563 B) ▸
    {"actor":"system:backfill","investigation_id":"96c6369e-4f55-49a7-8101-2c6d114057e2","kind":"publish","page_slug":"aave-v3","published_at":"2026-05-29T16:00:00.967Z","sequence_num":1,"snapshot":{"content_type":"investigation","entity_name":"Aave V3","sections":[{"content":"","heading":"","severity":"medium","sources":[{"credibility":3,"name":"https://aave.com/docs/aave-v3/overview","type":"other","url":""},{"credibility":3,"name":"https://coinlaw.io/aave-statistics/","type":"other","url":""},{"credibility":3,"name":"https://aave.com/blog/aave-2025-recap","type":"other","url":""},{"credibility":3,"name":"https://www.theblock.co/post/350525/how-aaves-stablecoin-gho-is-core-to-its-plan-to-10x-its-revenue","type":"other","url":""}]},{"content":"","heading":"","severity":"medium","sources":[{"credibility":3,"name":"https://governance.aave.com/t/rseth-incident-report-april-20-2026/24580","type":"other","url":""},{"credibility":3,"name":"https://www.coindesk.com/tech/2026/04/19/aave-records-usd6-billion-tvl-drop-as-kelp-hack-exposes-structural-risk-at-defi-lender","type":"other","url":""},{"credibility":3,"name":"https://news.bitcoin.com/zachxbt-flags-280m-kelpdao-exploit-hitting-ethereum-defi-lending-markets/","type":"other","url":""},{"credibility":3,"name":"https://unchainedcrypto.com/aaves-6-6-billion-tvl-drop-exposes-structural-risk-from-liquid-restaking-tokens/","type":"other","url":""},{"credibility":3,"name":"https://www.coindesk.com/tech/2026/04/20/aave-could-face-up-to-usd230-million-in-losses-after-kelp-dao-bridge-exploit-triggers-defi-chaos","type":"other","url":""}]},{"content":"","heading":"","severity":"medium","sources":[{"credibility":3,"name":"https://news.bitcoin.com/defi-united-secures-160m-as-industry-moves-to-cover-aave-bad-debt-crisis/","type":"other","url":""},{"credibility":3,"name":"https://www.coindesk.com/business/2026/04/26/aave-raises-nearly-80-of-the-usd200-million-it-needs-to-cover-bad-debt-left-by-kelp-dao-exploit","type":"other","url":""},{"credibility":3,"name":"https://www.coindesk.com/business/2026/04/23/aave-rallies-defi-partners-to-contain-fallout-from-usd292-million-kelpdao-hack","type":"other","url":""},{"credibility":3,"name":"https://unchainedcrypto.com/aaves-defi-united-recovery-effort-reaches-238-million-as-crypto-rallies-to-bail-out-its-biggest-lender/","type":"other","url":""},{"credibility":3,"name":"https://www.cryptotimes.io/2026/05/26/aave-and-kelp-dao-restore-rseth-operations-after-april-exploit/","type":"other","url":""},{"credibility":3,"name":"https://www.coindesk.com/markets/2026/05/18/aave-restores-weth-collateral-limits-as-rseth-crisis-enters-recovery-phase","type":"other","url":""},{"credibility":3,"name":"https://nftplazas.com/aave-commits-25000-eth-to-industry-wide-recovery-fund/","type":"other","url":""}]},{"content":"","heading":"","severity":"medium","sources":[{"credibility":3,"name":"https://governance.aave.com/t/aave-v2-v3-security-incident-04-11-2023/15335","type":"other","url":""},{"credibility":3,"name":"https://vibraniumaudits.com/post/aave-hacked-via-periphery-contract-56kstolenfromtipjar","type":"other","url":""},{"credibility":3,"name":"https://blog.solidityscan.com/aave-repay-adapter-hack-analysis-aafd234e15b9","type":"other","url":""},{"credibility":3,"name":"https://coinrabbit.io/blog/aave-hack-exploit-history-multi-million-losses-in-defi-crypto-hacks-and-key-lessons/","type":"other","url":""}]},{"content":"","heading":"","severity":"medium","sources":[{"credibility":3,"name":"https://aave.com/security","type":"other","url":""},{"credibility":3,"name":"https://oxor-io.github.io/public_audits/Aave/Aave-v3.3.0-Audit-Report.pdf","type":"other","url":""},{"credibility":3,"name":"https://www.certora.com/reports/aave-v3-risk-steward","type":"other","url":""},{"credibility":3,"name":"https://aave.com/docs/aave-v3/umbrella","type":"other","url":""},{"credibility":3,"name":"https://governance.aave.com/t/bgd-aave-safety-module-umbrella/18366","type":"other","url":""},{"credibility":3,"name":"https://blockworks.com/news/umbrella-reshapes-aave-staking","type":"other","url":""}]},{"content":"","heading":"","severity":"medium","sources":[{"credibility":3,"name":"https://www.ccn.com/news/crypto/sec-long-investigation-aave-everything-should-know/","type":"other","url":""},{"credibility":3,"name":"https://finance.yahoo.com/news/sec-closes-4-long-investigation-091715487.html","type":"other","url":""},{"credibility":3,"name":"https://www.mexc.com/news/290227","type":"other","url":""}]},{"content":"","heading":"","severity":"medium","sources":[{"credibility":3,"name":"https://governance.aave.com/t/rseth-incident-report-april-20-2026/24580","type":"other","url":""},{"credibility":3,"name":"https://news.bitcoin.com/incident-report-llamarisk-aave-service-providers-detail-kelp-rseth-hack-across-ethereum-and-arbitrum-markets/","type":"other","url":""},{"credibility":3,"name":"https://www.nydig.com/research/the-butterfly-effect-comes-to-defi","type":"other","url":""},{"credibility":3,"name":"https://thedefiant.io/news/defi/aave-price-crash-kelpdao-exploit-whale-dump-rxi8o9","type":"other","url":""}]}],"sources_used":[],"summary":"Aave V3 is the third major iteration of the Aave decentralized lending protocol, one of the largest in DeFi with over $14 billion in total value locked across 21 chains as of May 2026. The protocol's core smart contracts have not been directly exploited; however, in April 2026, a $292 million bridge exploit targeting integrated asset KelpDAO's rsETH generated significant bad debt on Aave V3 markets, an event flagged by on-chain investigator ZachXBT. The protocol demonstrated institutional resilience by coordinating a cross-industry recovery fund (DeFi United) that ultimately raised over $327 million, with full rsETH operations restored by May 25, 2026.","timeline":[{"date":"2022-03-16","event":"Aave V3 officially launched on Avalanche, Polygon, Arbitrum, Optimism, Fantom, and Harmony.","source":""},{"date":"2023-01-27","event":"Aave V3 deployed on Ethereum mainnet.","source":""},{"date":"2023-07-15","event":"GHO, Aave's native decentralized stablecoin, launched on Ethereum mainnet.","source":""},{"date":"2023-11-04","event":"Aave receives bug bounty disclosure of critical vulnerability affecting V2 pools. Guardian pauses V2 Ethereum and freezes assets on V2/V3 across multiple chains. No funds lost.","source":""},{"date":"2023-12-01","event":"All Aave V2/V3 markets resume full normal operations following November 2023 security incident.","source":""},{"date":"2024-06-20","event":"Certora completes formal verification and security audit of Aave V3 Risk-Steward contracts.","source":""},{"date":"2024-08-28","event":"ParaSwapRepayAdapter periphery contract exploited for approximately $56,000. Core Aave V3 contracts unaffected.","source":""},{"date":"2025-01-29","event":"Oxor completes security audit of Aave V3.3.0 smart contracts.","source":""},{"date":"2025-06-01","event":"Aave Umbrella safety module launches on Ethereum with support for aUSDC, aUSDT, aWETH, and GHO staking.","source":""},{"date":"2025-12-16","event":"SEC closes four-year investigation into Aave with no enforcement action.","source":""},{"date":"2026-04-18","event":"KelpDAO rsETH LayerZero bridge exploit: attacker mints 116,500 unbacked rsETH and deposits 89,567 rsETH as collateral on Aave V3, borrowing ~$191 million. ZachXBT flags the exploit publicly.","source":""},{"date":"2026-04-18","event":"Aave Protocol Guardian freezes rsETH/wrsETH reserves across all 11 affected V3 deployments within hours of the exploit.","source":""},{"date":"2026-04-19","event":"Aave TVL falls approximately $6.6 billion. AAVE token declines 10-13%.","source":""},{"date":"2026-04-20","event":"Aave service providers publish rsETH Incident Report, estimating bad debt between $123.7 million and $230.1 million.","source":""},{"date":"2026-04-23","event":"Aave launches DeFi United coalition to raise funds covering bad debt. Aave DAO and Mantle pledge combined 55,000 ETH.","source":""},{"date":"2026-04-25","event":"DeFi United raises $160 million, representing approximately 80% of estimated bad debt coverage needed.","source":""},{"date":"2026-05-18","event":"Aave restores normal WETH loan-to-value ratios across six V3 networks as recovery exceeds 95%.","source":""},{"date":"2026-05-25","event":"Aave and Kelp DAO announce full restoration of rsETH operations. DeFi United total raised surpasses $327 million. Approximately 106,993 of 112,103 unbacked rsETH recovered or offset.","source":""}]},"v":1}
    Verify offline (run on your own machine)
    python -m src.verify_decision 9c8963e9-bf6e-4c7c-97fc-0b6d315e189d
  2. #2reviewby reviewerreviewer
    2026-06-04 03:24:23Z
    Score: 6868 (no score change)
    The Aave V3 investigation page is largely accurate on core facts — the exploit mechanics, bad debt estimates, and timeline dates are well-supported by independent sources. The most significant verifiable errors are: (1) the AAVE token decline is understated at 10-13% when primary sources report 16-20%; (2) the Umbrella launch date is June 5 not June 1; (3) the DeFi United total of $327M is not corroborated — credible sources cite $300M-$320M; (4) the specific rsETH recovery figures (106,993/112,103) do not match the 116,131.72 rsETH restoration reported in primary sources. All section 'content' fields are empty, meaning the investigation has no body text, limiting its standalone utility.
    anchoranchored
    chain
    mainnet-betaslot 424,161,624
    sig
    5aH4XSB3pBbK…YKTPnjVZexplorer ↗
    hash
    8j9DZEtERR4c…eC9fQbCtsha256 → base58
    verifying row…full verify ↗
    canonical bytes (1039 B) ▸
    {"actor":"reviewer","decided_at":"2026-06-04T03:24:23.544Z","decision":"review","investigation_id":"96c6369e-4f55-49a7-8101-2c6d114057e2","new_score":68,"page_slug":"aave-v3","prev_score":68,"reason":"The Aave V3 investigation page is largely accurate on core facts — the exploit mechanics, bad debt estimates, and timeline dates are well-supported by independent sources. The most significant verifiable errors are: (1) the AAVE token decline is understated at 10-13% when primary sources report 16-20%; (2) the Umbrella launch date is June 5 not June 1; (3) the DeFi United total of $327M is not corroborated — credible sources cite $300M-$320M; (4) the specific rsETH recovery figures (106,993/112,103) do not match the 116,131.72 rsETH restoration reported in primary sources. All section 'content' fields are empty, meaning the investigation has no body text, limiting its standalone utility.","score_delta":0,"sequence_num":2,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}
    Verify offline (run on your own machine)
    python -m src.verify_decision 14c8bc64-8947-493a-992c-932a6ec99f6f
  3. #3review reviseby judgejudge
    2026-06-04 03:24:23Z
    Score: 6858 (-10)
    The reviewer confirmed 13 of 20 claims outright with strong tier-1 sourcing. The disputed_pct of 15% places this page in the revise band. The single hard-disputed finding (claim_findings[12]: Umbrella launch date stated as June 1, confirmed as June 5 by multiple independent sources) is a minor date error. More consequentially, the $327 million DeFi United fundraise total — stated in both the summary and claim_findings[17] — is not corroborated by any source; credible reporting consistently cites $300M-$320M, making this a repeated overstatement of a headline recovery figure. The AAVE token decline in claim_findings[12] is also understated at 10-13% when primary sources report 16-20%, which affects readers' assessment of market impact severity. Additionally, all seven section content fields are empty (coverage_gaps, priority: high), meaning the page has no body text beyond the summary and timeline — a structural deficiency requiring revision before the page can stand alone as a useful investigation.
    anchoranchored
    chain
    mainnet-betaslot 424,161,627
    sig
    5pphkK31syU9…x7KBeoz4explorer ↗
    hash
    ANLx5dR5UGK4…rEoMhCcDsha256 → base58
    verifying row…full verify ↗
    canonical bytes (1362 B) ▸
    {"actor":"judge","decided_at":"2026-06-04T03:24:23.544Z","decision":"review_revise","investigation_id":"96c6369e-4f55-49a7-8101-2c6d114057e2","new_score":58,"page_slug":"aave-v3","prev_score":68,"reason":"The reviewer confirmed 13 of 20 claims outright with strong tier-1 sourcing. The disputed_pct of 15% places this page in the revise band. The single hard-disputed finding (claim_findings[12]: Umbrella launch date stated as June 1, confirmed as June 5 by multiple independent sources) is a minor date error. More consequentially, the $327 million DeFi United fundraise total — stated in both the summary and claim_findings[17] — is not corroborated by any source; credible reporting consistently cites $300M-$320M, making this a repeated overstatement of a headline recovery figure. The AAVE token decline in claim_findings[12] is also understated at 10-13% when primary sources report 16-20%, which affects readers' assessment of market impact severity. Additionally, all seven section content fields are empty (coverage_gaps, priority: high), meaning the page has no body text beyond the summary and timeline — a structural deficiency requiring revision before the page can stand alone as a useful investigation.","score_delta":-10,"sequence_num":3,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}
    Verify offline (run on your own machine)
    python -m src.verify_decision 2569f575-5a4a-4085-b651-9b115eca8bac
How verification works. The “Row integrity” check above is computed in your browser — your machine recomputes the SHA-256 of the canonical bytes and compares against the stored hash. No avoid.net server can fake that check. The “full verify” link goes one level deeper: your browser fetches the on-chain transaction from a Solana RPC node and confirms the same hash is in the memo. If you don’t want to trust either avoid.net or the public RPC, run the CLI verifier on your own machine — python -m src.verify_decision <event_id>.