What is Composability Risk?

1 min read Updated

Composability risk is the danger that arises when multiple DeFi protocols interact in ways that create unexpected vulnerabilities, cascading failures, or economic exploits — a consequence of DeFi's building-block architecture where protocols build on each other.

WHY IT MATTERS

DeFi's greatest strength is composability — protocols can be combined like Lego blocks. Deposit collateral in Aave, borrow against it, swap on Uniswap, provide liquidity in Curve. This composability creates powerful financial products but also systemic risk.

When protocols are interconnected, a failure in one can cascade. If an oracle feeding prices to Aave fails, it affects not just Aave but every protocol that depends on Aave positions (yield aggregators, automated strategies, leveraged positions). The interconnection amplifies risk.

AI agents operating across multiple protocols amplify composability risk further. An agent simultaneously managing positions in Aave, Uniswap, and Curve has exposure to all three protocols AND the interactions between them. A composability exploit could affect all positions simultaneously.

HOW POLICYLAYER USES THIS

PolicyLayer limits agent exposure to composability risk by restricting which protocols agents can interact with, limiting cross-protocol positions, and enforcing diversification requirements. This bounds the blast radius of composability failures.

FREQUENTLY ASKED QUESTIONS

What's an example of composability risk?
The Cream Finance exploit: an attacker used flash loans across multiple protocols (Cream, Aave, Curve) to manipulate prices and drain funds. No single protocol was vulnerable alone — the exploit was only possible through their combination.
How do you mitigate composability risk?
Limit exposure per protocol, avoid excessive leverage across protocols, monitor oracle dependencies, diversify across independent protocol stacks, and set position limits that account for correlated risks.
Does composability risk increase with more protocols?
Yes. Each additional protocol in a composition adds risk. The relationship is often non-linear — three interconnected protocols can have more risk than the sum of their individual risks.

FURTHER READING

Enforce policies on every tool call

Intercept is the open-source MCP proxy that enforces YAML policies on AI agent tool calls. No code changes needed.

npx -y @policylayer/intercept
github.com/policylayer/intercept →
// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.