rate-limiting

Rate limiting restricts the number of requests a client can make within a time window, preventing abuse and protecting service stability. Token bucket, sliding window, and leaky bucket are common algorithms; this Skill implements all three with Redis-backed state for distributed systems, per-client quotas, and burst handling.

Distributed rate limiting uses Redis-backed state to enforce quotas consistently across multiple services and API gateways. This Skill provides patterns for per-IP controls, per-user quotas, and backpressure signaling, enabling fair usage policies in multi-tenant environments without duplicating state.

Yes. Redis provides fast, shared state management for rate-limiting algorithms across microservices. This Skill implements Redis-backed sliding windows and token bucket patterns, allowing you to maintain consistent quotas and reset times across distributed API gateways and multi-tenant deployments.

Token bucket allows burst traffic up to a capacity limit before throttling; sliding window tracks request counts over a rolling time period for stricter enforcement. This Skill implements both, letting you choose based on whether your use case prioritizes burst flexibility or precise request distribution.

Rate-limiting observability includes remaining quota, reset time, and backpressure signals, helping clients understand throttling status. This Skill provides these metrics for burst handling and fair client behavior, enabling API consumers to implement smart retry logic and request scheduling.

Backpressure prevents cascading failures when upstream services reject requests due to rate limits. This Skill includes backpressure patterns alongside per-client quotas and reset-time signaling, critical for stable multi-service architectures and preventing thundering-herd effects.