A vault can compress many contracts and decisions into one share token. That convenience does not reduce the assessment to a displayed rate or a curator label.
The useful question is whether the product's authority, positions, dependencies, exits, and evidence form a system you understand well enough to use.
The object of assessment
The object is not the curator in isolation. It is the product formed by contracts, permissions, assets, venues, operators, and the rules that connect them.
The vault holds positions and defines enforceable permissions. The curator frames or monitors an agreed mandate. The depositor decides whether that system fits the decision at hand.
Those roles can differ by product. Start with the vault's own contracts and documentation before assigning authority from a category label.
Start with the mandate
A mandate states what the product may do. Read eligible assets, venues, allocation limits, liquidity expectations, fees, and the conditions for changing those terms.
Separate objective from permission. A stated aim describes intent; the contracts and governance path determine which actions are possible.
Then ask who can allocate, change limits, pause activity, or alter the product. An assessment without an authority map leaves the most important transition unexplained.
Read positions against permission
Current positions show what the vault holds now. The mandate shows what it could hold later. Neither view is sufficient on its own.
Compare each position with the permitted set. Note unused permissions, temporary balances, borrowed exposure, maturity, and any position whose classification hides a material dependency.
A snapshot is time-bound. Check its observed-at time and source before treating it as the product's present state.
Map dependencies and concentration
Positions that look different can share the same protocol, token, oracle, bridge, issuer, governance process, or liquidity venue.
Aggregate those shared dependencies before judging diversification. Several line items may still express one failure mode.
Concentration is contextual. A weight matters alongside collateral quality, liquidity, correlation, control rights, and the cost of reducing the exposure.
Test the exit path
An exit is a sequence, not a button. Trace how positions unwind, what liquidity they require, which maturities or queues apply, and who can initiate each step.
Ask what changes under stress. Slippage, utilisation, redemption delays, oracle movement, or a paused venue can make the normal route irrelevant.
Do not turn an observed withdrawal into an assurance. The current path can change before a depositor acts.
Treat the evidence boundary as part of the product
Every material observation needs a source and a time. Rates also need a period, method, and fee or compounding treatment.
Unavailable evidence is not neutral. It limits what can be concluded about a position, fee, rate, dependency, or exit.
Use the public dashboard as a reading surface, then follow its links to the underlying contracts and source systems. The presentation layer is not the product's source of truth.
Reassess after deposit
Assessment does not end when capital enters. Positions, utilisation, incentives, liquidity, governance, and available evidence can all change.
Define review triggers before they are needed. A mandate change, stale source, new dependency, concentration breach, or impaired exit can justify a fresh decision.
Tulipa's public methodology describes the shared review frame. It supports diligence; it does not replace product-specific analysis or individual judgment.