Katana | Infrastructure Partners — Oracles, Curators & Providers
Understand the infrastructure roles that support Katana, including RPC providers, bridges, curators, oracles, security participants and application partners.
Why partners matter
Katana is a network, but many user-facing services depend on independent infrastructure. RPCs, bridges, curators, oracle providers, indexers, security participants and core apps each introduce their own operational and risk assumptions.
RPC and node providers
Applications rely on RPC providers to read state and submit transactions. Provider outages or rate limits can make an interface appear broken even when the chain is healthy. Production systems should support provider failover.
Curators and vault managers
Morpho and other vault strategies can be managed by independent curators that select markets and risk limits. Curator quality affects strategy behavior but does not transfer custody or governance of Katana itself.
Oracles and data services
Lending markets and DeFi strategies depend on reliable asset pricing. Oracle design, update cadence and failure modes are especially important for collateralized markets where bad data can trigger liquidations.
Security participants
Katana’s Security Council includes ecosystem and infrastructure participants that provide oversight and emergency response capabilities. This is separate from the day-to-day operation of third-party apps.
Integration rule
Treat every external dependency as an explicit component. Document what happens if it is unavailable, stale or compromised, and avoid assuming that a Katana integration automatically inherits the security of the base network.
Frequently Asked Questions
Are all Katana apps controlled by Katana?
No. Core and integrated apps can be independent protocols deployed on Katana.
Why do RPC providers matter?
Apps need RPC access to read chain state and submit transactions.
What do vault curators do?
They select markets and risk parameters for curated strategies.
Should developers plan for provider failure?
Yes. Resilient applications should handle outages, stale data and provider failover.