Lesson 8 — Pre-deposit checklist: the questions to ask before any DeFi allocation

**1. TVL and TVL history.** Look up the protocol on DeFiLlama. What is current TVL? What is TVL six months ago, one year ago, two years ago? Steeply declining TVL is a leading indicator — often the people closest to the protocol have already exited. Conversely, sudden very-fast TVL growth without organic explanation is sometimes a points-farming dynamic that will reverse equally fast. Stable, slowly-growing TVL over multiple years on a heavily-audited protocol is a strong baseline.

**2. Audit history.** How many audits, from which firms, on which versions? Are the audits on the actual deployed code (commit hashes match Etherscan-verified source)? Are there unresolved Critical or High findings? Is the audit recent enough that material code changes since are minimal? Skip protocols with a single audit older than a year on code that has been substantively changed.

**3. Admin functions and timelocks.** Read the protocol's governance and admin documentation. Who can pause the contracts? Who can upgrade? Who can change critical parameters (interest rate curves, collateral factors, oracle sources)? Is there a timelock between proposal and execution? A 48-hour timelock on every admin action is the minimum that gives users time to exit if a malicious proposal is queued. A protocol where a single multisig can change parameters with no timelock is functionally a custodial product — the team can change the rules at any moment.

**4. Oracle design.** Which oracle, which feed, which update frequency? Is the oracle one of the well-tested feeds (Chainlink mainline, Pyth on Solana for major assets) or a long-tail feed with less venue aggregation? For any protocol where prices affect consequential operations (liquidations, settlements, redemptions), the oracle is the load-bearing dependency. Lesson 5's pattern recognition applies directly here.

**5. Withdrawal limits and circuit breakers.** Does the protocol have a withdrawal-rate limit that would slow a drain during a stress event? Is there an emergency pause? Who controls the pause? Protocols with no rate limits and no pause functionality have no circuit breaker — when a drain starts, it runs to completion. The trade-off is that circuit breakers add admin power, so they need to be coupled with timelocks and multisig controls.

**6. Bridge dependency.** Does this protocol's token live natively on this chain, or is it a bridged wrapper? If it's bridged, which bridge? Apply Lesson 6's framework: signer-set distribution, operational history, key custody. Holding 'USDC' on a chain via an unofficial bridge stacks the bridge risk on top of the issuer risk. Holding native USDC (Circle's direct deployment) on chains where Circle supports it is a clean stack.

**7. Team transparency.** Are the team members public? Are they reachable? Have they been involved in previous DeFi projects? Anonymous teams are not categorically disqualifying — some excellent protocols have anonymous founders — but the operational track record needs to be longer and the technical openness needs to be greater to compensate. A team that has been public, named, and reachable across several years builds reputational capital that anonymous teams structurally cannot.

**8. Documentation and code-to-doc match.** Read the protocol's documentation. Does it accurately describe what the contracts do? Are the contracts open-source and verified on Etherscan? Can you trace any given user flow from the docs to the actual code? Protocols whose documentation doesn't match their code, or whose contracts aren't verified, are exposing users to obfuscation risk regardless of the documentation's quality.

**9. Insurance and post-incident track record.** Does the protocol have insurance coverage available (Nexus Mutual, InsurAce, others)? Have past incidents at this protocol been disclosed and compensated, or quietly buried? A protocol whose past minor issues were openly disclosed and addressed is signaling that future incidents will be handled the same way. A protocol whose history is suspiciously clean often has incidents that were hushed.

**10. Size your position to your risk tolerance.** No checklist eliminates risk. Major exploits have hit protocols passing every checklist item. The final discipline is sizing positions so that a total loss of any single protocol position is survivable. For most users that means no single protocol holds more than 5–15 percent of total DeFi exposure, with the most-mature protocols (Aave, Compound, Maker, Lido, Curve, Uniswap) earning the higher end of the range.

**Course closing.** Eight lessons in, you have the toolkit: AMM math, lending mechanics, audit reading, exploit-class pattern recognition, oracle evaluation, bridge analysis, IL/LVR math, and this checklist. The toolkit doesn't tell you what to allocate to. It tells you what questions to ask before allocating. Applied consistently to any new protocol — including the next one that will be exploited — it would have prevented a substantial fraction of historical losses. The remaining losses come from risks the toolkit doesn't fully cover (unknown unknowns, sophisticated novel exploits, integration surprises), and that residual is what position sizing is for.