agentleFS
Maxrall-IncMaxrall-Inc/nexrall-skills/skills/defi-risk-assessment/SKILL.md

defi-risk-assessment

Skill0 starsChanged 3 months agoMITView on GitHub

Assess the risk of a DeFi protocol or position across smart-contract risk, liquidity, oracle dependence, and concentration, before the user deposits or invests. Use when the user asks "is this protocol safe", "should I put money in this", or wants a yield pool, lending market, or token evaluated.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

DeFi Risk Assessment

DeFi risk is not one number — it is a stack of independent failure modes, and a protocol that is safe against one can be wide open on another. A user deciding whether to deposit needs the separate risks named, sized, and weighted, not collapsed into a single "safe / not safe" verdict that hides which assumption would have to break.

The risk stack (assess each independently)

1. Smart-contract risk

Can the code lose funds on its own? This is the domain of the smart-contract-review skill — run that lens first: reentrancy, access control, unchecked calls, upgradeability (is the proxy admin a multisig or a single EOA?). For an established protocol, note whether it has public audits and a bug bounty; for a new one, the absence of both is itself a finding.

2. Liquidity risk

Can you get out?

  • Depth: how much can be withdrawn/swapped before meaningful slippage?
  • Lockup: is there a staking lock, a vest, or a withdrawal delay?
  • Exit: is there a single liquidity provider / pool that, if pulled, strands everyone else?

3. Oracle / price risk

What price does the protocol trust, and can it be moved?

  • Spot-DEX prices are flash-loan manipulable within one block.
  • A stale or centralized oracle (a single reporter) is a single point of failure.
  • A token whose price the protocol derives from its own pool is circular and exploitable.

4. Concentration / governance risk

  • Token supply: do a few addresses control governance or the majority of the supply?
  • Admin keys: can an owner mint, pause, or upgrade unilaterally? (A protocol with an active admin key is "trust me", not "trustless", no matter the marketing.)
  • Upgradeability: what changes can a governance vote or multisig push, and how quickly?

5. Counterparty / composability risk

What does the protocol depend on?

  • Other protocols (a lending market built on another lending market inherits that one's risk).
  • Bridges and wrapped assets (a bridged token is only as good as the bridge).
  • Stablecoins with their own solvency question.

Procedure

  1. Identify the protocol and the specific position the user is considering (which pool, which asset, how much, for how long).
  2. Gather evidence for each of the five risks — contracts, audits, TVL, liquidity depth, holder concentration, oracle design.
  3. Rate each risk Low / Medium / High, with a one-line reason.
  4. Give the verdict in terms of what has to be true for you to lose money, not a single score.

Output format

Protocol / position: <name>
- Smart-contract risk: Low/Med/High — <reason>
- Liquidity risk:    Low/Med/High — <reason>
- Oracle/price risk: Low/Med/High — <reason>
- Concentration/governance risk: Low/Med/High — <reason>
- Composability risk: Low/Med/High — <reason>
Biggest single risk: <the one failure mode that dominates>
Verdict: <a sentence that names the assumption the position depends on>

Boundaries

This is an assessment of technical risk, not financial advice and not a guarantee. State that explicitly. A "Low" rating is still not "risk-free" — crypto has no risk-free yield. If an offered yield is far above the market average, lead with the question that explains most DeFi losses: where is the yield actually coming from? — and check whether it is just paying early depositors from new deposits (a ponzi dynamic) rather than real protocol revenue.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.