No SLA or fallback standard exists for keeper execution. Missed triggers leave users with no recourse.
Pillar: Autonomy
The problem
Keeper networks (Gelato, Keep3r, Chainlink Automation) are the execution layer for automated DeFi strategies. They have no binding service-level agreement. If a trigger is missed, the user has no notification, no fallback, and no recourse.
The reliability of automation is entirely opaque to the user who depends on it.
Why it matters
- Autonomy: The execution layer I depend on offers no guarantees.
- Safety: Missed liquidation protection triggers are a direct path to position loss.
What exists today
Keeper networks publish uptime metrics. No standardized SLA, fallback, or user notification framework.
The gap
No standard for keeper reliability disclosure, fallback execution, and failure notification.
Open questions
- Can redundant keeper registration solve this at the application layer?
- What would a meaningful SLA look like for decentralized keepers?