Questions

List of Questions

Security

Smart Contract Security

Whats the risk of an on-chain protocol exploit?

Have all core contracts that hold or route funds been independently audited?

Who performed the most recent audit and what is their reputation tier?

Have upgrades to core contracts been followed by new audits?

Have all critical and high-severity audit findings been fully remediated and re-verified?

Are core contracts upgradeable and how tightly are upgrades constrained?

Has any on-chain exploit or critical bug affecting user funds occurred, were affected users compensated and were post-mortems published?

Has any confirmed rug-pull event occurred?

Were there reductions to the timelock delay that weaken governance protections and if so, have they been communicated transparently?

Does the protocol run an active bug-bounty via a reputable platform (e.g. Immunefi, Code4rena)?

Are real-time security monitoring and alerting systems in place?

Are automatic safety controls (e.g. pause, circuit breakers) triggered by monitoring alerts?

Key Management & Permissions

What is the risk of an off-chain protocol exploit?

Who controls admin and upgrade keys for (proxy-) contracts holding user funds and have previous changes been documented?

Can any single key move user funds or upgrade custody contracts?

How decentralized and robust is the multisig for critical actions and have modficiations to the signing configuration been documented?

How constrained are pause, blocklist and withdrawal-control permissions?

Are all user assets held in non-custodial smart contracts?

Are user funds fully segregated from treasury and operational wallets?

What are the whitelisted protocols the vault strategy can interact with, how restirctive is it to the mandate and how are modifications permissioned?

Is there a tested incident playbook for admin-key compromise or signer loss?

Strategy

Protocol Mechanics

How likely does the yield source goes negative?

What are the sources of yield aligned with the mandate of the vault or protocol?

Under which conditions does net yield become negative or cause loss?

Is leverage or rehypothecation used and up to what effective level?

Are there automated mechanics to prevent Liquidation Events (health rate, LTV thresholds)

Has this strategy or a close variant ever experienced negative yield?

Is a risk framework for vault curation or protocol mechanics documented?

Is current yield sustainable relative to underlying economics?

Can the position be fully or mostly unwound under conservative liquidity assumptions without slippage, penalties, or lockups turning expected yield negative?

Does the strategy rely on a spread, peg, or funding rate that can invert (e.g. basis trades, stablecoin pegs, LST discounts, market making)?

Collateral

How likely is a collateral depeg?

Which assets are accepted as collateral and how are they risk rated?

How did each collateral behave versus its underlying during past stress or depegs?

How concentrated is exposure to any single collateral or LST?

How is validator or slashing risk handled for staking-based collaterals?

What share of TVL relies on bridged or wrapped assets?

Infra Counterparty Exposures

How likely does a malfunction of counterparty Chain, Bridge, Oracle, Wallet, Validators, APIs, Frontends or CEX affect the principal?

Which chains, bridges, oracles, wallets and CEXs does the strategy depend on and how are they risk rated or tiered?

How redundant and battle-tested are the oracle and bridge setups?

Are off-chain infrastructure providers certified by standard IT security audits like SOC 2 or ISO27001 or VSP?

How did these infra components behave in past outages or chain incidents?

Has the base chain recently halted block production, experienced consensus failure or become non-operational?

Has the validator set experienced slashing events that could impact staked collateral?

Are validators diverse geographically and by operator?

Can any single infra component failure alone block withdrawals or cause losses?

Protocol Counterparty Exposures

How likely does a malfunction of counterparty protocol affect the collateral?

Into which external protocols, CEXs or RWA platforms is collateral deployed and how are they risk rated?

Are revenue-share or incentive arrangements documented and do they create conflicts of interest?

Are there hard limits and rebalancing rules per counterparty?

How quickly and safely can positions be unwound from each counterparty?

Are third-party validators, custody providers or asset managers disclosed?

Liquidity

How likely is it that 100% of the collateral can be withdrawn within the stated withdrawal period?

How are withdrawals executed and are they enforced permissionlessly on-chain or can they be paused, blocked or delayed?

Is there enough liquidity and market depth to redeem close to 100% of TVL within the stated withdrawal time?

Have there been past withdrawal delays queues or freezes and does the avg processing time match expectations?

Do caps or limits restrict timely redemptions during high utilization?

How does exit liquidity behave during volatility and network congestion?

Do withdrawals rely on unstaking, vesting, or bridge exits longer than the stated period?

Does redemption depend on secondary-market liquidity and is it sufficient?

What is the liquidity depth of collateral and receipt tokens?

What is the risk of bank-run scenarios under full utilization?

Market

What happens in a market black-swan event or high market volatility?

How has the LST or receipt token peg performed under market stress?

How likely can adverse moves in broader crypto markets cause permanent losses due to directional exposure?

Are interest rate/incentive models robust during TVL shifts or volatility spikes?

Have historic scenarios with high market volatility been explicitly modeled for the strategy?

Can sharp price moves trigger forced liquidations that crystallize losses before positions can be adjusted?

Operations

Governance

How likely can governance changes or capture cause losses or block withdrawals?

What governance model controls protocol changes and upgrades?

How concentrated is voting power among top holders or delegates and how active transparent and user-aligned is governance in practice?

Are mechanisms in place to limit major changes in scope and time like timelocks and veto or delays?

Can the protocol replace the strategy manager without blocking user withdrawals and how is this governed?

How robust is the team and legal setup in protecting my collateral from fraud or seizure?

Are core team and operating entities publicly identified and credible?

Is the protocol dependent on a single developer or a small team, and does this ensure optimal management for the strategy?

What legal entity and jurisdiction operate the protocol and/or assets?

Are there known investigations or regulatory actions related to the entity?

Is there an on-call and incident response process for core teams and strategy managers?

Does the team provide timely support for critical user or integrator issues?

Are major investors or strategic partners disclosed?

Does the strategy manager manage similar vaults and does it introduce a conflict of interest introducing sub-optimal risk descisions?

Has the strategy manager previously been involved in products with collateral loss or materially adverse curation decisions?

Documentation & Transparency

How clearly does the protocol explain what happens to my collateral and when I could lose it?

Is there up-to-date documentation explaining architecture, strategy mechanics, and risks?

Is it clearly documented how collateral is used and when loss occurs?

Are all contract, proxy and implementation addresses documented in one canonical place and verified on-chain?

Are roles, permissions and timelocks documented with a visible change history?

Does the protocol publish proof-of-reserves and where applicable proof-of-liabilities?

Are all contracts verified on the leading block explorers?

Financial Resilience

How likely can the protocol absorb shocks without passing permanent losses to depositors?

Is there backstop reserve or safety module for user losses and is there any guarantee it will be used in a loss event?

How large and liquid are the backstop reserves and treasury relative to TVL and are they held in low-risk, liquid assets?

What is the estimated operational runway at current burn for operational teams and managers?

How have TVL revenue and buffers behaved in past stress events?

Can the protocol remain safe in maintenance mode if team disappears?

Resources

Excel Sheet with questions: GitHubarrow-up-right.

Last updated