KVORUM
aaveSource: Aave governance v3 · #389

Adopt The SEAL Safe Harbor Agreement

executedOct 14, 2025, 07:17 AM UTCbyF70xf71f…1e02

01Summary

AISummary by Kvorumclaude-haiku-4-5

Aave adopts the SEAL Whitehat Safe Harbor Agreement on-chain to enable legal protection and swift intervention by security researchers during active exploits. The agreement provides a 10% bounty (capped at $1M per incident) for recovered funds, covers all Aave V3 pools and infrastructure across 16 chains, and is funded from the DAO treasury rather than recovered user funds.

Type
governance change
Funding
$1000000

Key changes

  • Registration of Aave's adoption of SEAL Safe Harbor Agreement providing legal protection for whitehats intervening during active exploits(high)
  • Establishment of 10% bounty structure (capped at $1M per incident, $1M aggregate) for successful fund recovery by whitehats(high)
  • Coverage extended across all Aave V3 pools (Core and Prime), governance infrastructure, GHO token infrastructure, and Umbrella contracts across 16 chains(high)
  • Clarification that Safe Harbor and Bug Bounty rewards are mutually exclusive for the same exploit(medium)
  • Bounty payments sourced from Aave DAO treasury with flexible reward denomination at DAO discretion(medium)

Notable concerns

  • KYC and sanctions screening required when necessary, potentially affecting whitehat anonymity
  • Bounty cap of $1M may be insufficient for very large exploits relative to recovered funds
  • Safe Harbor and Bug Bounty mutual exclusivity may create coordination issues if multiple response mechanisms are applicable
Contracts0x1eaCD100B0546E433fbf4d773109cAD482c346860x276c889f5312844c6d3Df9Ed6C742649f79F9a9DAll Aave V3 pools and components across 16 chains
Generated by Kvorum
model: claude-haiku-4-5
provider: anthropic
profile: anthropic-fast-v1
prompt: v1.0
generated: 2mo ago
Read the description ↗

02Description

468 words

Simple Summary

This proposal registers Aave's adoption of the SEAL (Security Alliance) Whitehat Safe Harbor Agreement on-chain. The agreement enables whitehats to legally intervene during active exploits to protect protocol funds, with clear rules for fund recovery, predefined bounties, and legal protections for good-faith actors.

Motivation

While Aave maintains strong security through audits and preventive measures, the unpredictable nature of exploits requires swift response mechanisms. The Safe Harbor Agreement addresses a critical gap: whitehats often hesitate to intervene during active exploits due to legal uncertainty, even when they could save funds. By providing legal protection and clear operating guidelines, this agreement removes barriers that prevent whitehats from acting in good faith.

Key benefits include:

  • Rapid Response: Whitehats can intervene immediately during active exploits, minimizing damage without requiring protocol halts
  • Clear Process: Predetermined recovery addresses and procedures eliminate chaotic negotiations during crises
  • Fair Incentives: A 10% bounty (capped at $1M) aligns with Aave's existing bug bounty program, incentivizing protection without creating conflicting priorities
  • Legal Clarity: Provides whitehats with legal protection when acting in good faith, encouraging intervention over inaction
  • Industry Alignment: Positions Aave at the forefront of DeFi security practices

The agreement complements existing security measures by providing an additional defensive layer specifically for active threat scenarios.

Specification

The proposal executes a single transaction to register Aave's adoption in the Safe Harbor Registry:

Transaction Details:

Adoption Parameters:

  1. Asset Recovery Addresses: Multi-chain collector addresses controlled by Aave (see ARFC for complete list across 16 chains)
  2. Scope: Comprehensive coverage including:
    • All Aave V3 pools (Core and Prime on Ethereum, plus 15 other chains)
    • All pool components (configurators, oracles, adapters, gateways)
    • Governance infrastructure (cross-chain controllers, payload controllers, executors)
    • GHO token infrastructure across multiple chains
    • Umbrella, treasury, and v2 contracts
    • Contact: BGD Labs ([email protected])
  3. Bounty Terms:
    • Percentage: 10% of recovered funds
    • Cap: $1,000,000 USD per incident
    • Aggregate Cap: $1,000,000 USD
    • Retainable: False (all funds returned to protocol, bounty paid separately from DAO treasury)
    • Identity: Named (full legal name may be required)
    • Diligence: KYC & sanctions screening only when necessary

Important Notes:

  • Bounty payments come from Aave DAO treasury, not recovered user funds
  • Safe Harbor and Bug Bounty rewards are mutually exclusive for the same exploit
  • KYC requested sparingly, respecting whitehat anonymity when possible
  • Reward denomination at DAO discretion, volatile assets priced using 30-day average from incident date

References

Copyright

Copyright and related rights waived via CC0.

03Mismatch analysis

AIMismatch analysis by Kvorummedium confidenceclaude-sonnet-5
Minor discrepancy

The proposal describes a single, specific action: calling adoptSafeHarbor(address) on the Safe Harbor Registry, passing an Adoption Details contract address that encodes recovery addresses, scope, and bounty terms. The decoded calldata shows only one action: a call to execute() (a fallback-style entry point) on a different contract address, consistent with Aave's standard governance payload pattern where an on-chain payload contract is invoked and then internally executes the specific logic (here, presumably calling adoptSafeHarbor on the registry with the correct adoption details address). This pattern is routine for Aave V3 governance proposals and is not inherently suspicious, but because only the top-level execute() call is decoded, the actual downstream call to the registry and the exact adoption details address cannot be confirmed from the data given. I could not verify the specific bounty terms, recovery addresses, or scope details described in the proposal against the calldata, since none of that granularity appears in the decoded action. Given the widely recognized executor/payload pattern for Aave DAO proposals, I do not consider this a material or severe issue, but flag it as a minor discrepancy due to the inability to fully verify the internal call matches the stated specification from the calldata alone.

Discrepancies

  • lowtarget mismatch

    The description states the transaction target is the Safe Harbor Registry (0x1eaCD100B0546E433fbf4d773109cAD482c34686) with function adoptSafeHarbor(address). However, the decoded on-chain action shows a call to execute() on a different contract (0xbbfa2490332f93e33693dee1933d2d5ae8b64d80), which is presumably a payload/executor contract implementing the described logic internally. This is a common and expected pattern in Aave's governance flow (proposal payload calls the registry internally), but since only the top-level execute() call is decoded here, the actual internal call to adoptSafeHarbor cannot be independently verified from the provided calldata.

    “To: 0x1eaCD100B0546E433fbf4d773109cAD482c34686 (Safe Harbor Registry) ... Action: adoptSafeHarbor(address)”

Generated by Kvorum
model: claude-sonnet-5
provider: anthropic
profile: anthropic-strong-v1
prompt: v1.0
generated: 2mo ago
Review the decoded actions ↗

04Actions

1 total
#0fallback()chain 0x1
Target
0xbbfa…4d80
Value
0 wei
Function
fallback()
Arguments
{}
Calldata

05Tally

Summed from votes

Voters44
VP participating846.9K
Outcome (current)For
Current support100%

06Voters

44 total

No votes recorded yet.

07Forum

No forum thread is confidently linked to this proposal.