AgentSkillsCN

redis-caching-patterns

针对应用性能的Redis缓存工作流程。当必须实现或修订特定于框架的应用结构与运行时行为时使用；切勿用于整个代码库的架构治理或发布管理策略。

SKILL.md
--- frontmatter
name: redis-caching-patterns
description: "Redis caching workflow for application performance. Use when framework-specific application structure and runtime behavior must be implemented or revised; do not use for repository-wide architecture governance or release management policy."

Redis Caching Patterns

Trigger Boundary

  • •Use when Redis is introduced or tuned for cache/session/queue-like workloads.
  • •Do not use for primary relational schema design.
  • •Do not use for in-memory caching that does not involve Redis.

Goal

Improve latency and throughput with explicit cache correctness and invalidation policy.

Inputs

  • •Target read/write hotspots and latency objectives
  • •Data freshness and consistency requirements
  • •Eviction policy and memory constraints

Outputs

  • •Key namespace and TTL strategy
  • •Cache invalidation and degradation behavior plan
  • •Observability checklist for hit rate, memory, and evictions

Workflow

  1. •Identify cache candidates by read amplification and recomputation cost.
  2. •Design key schema with clear namespace and versioning.
  3. •Define TTL, invalidation triggers, and stale-read tolerance.
  4. •Add protection against cache stampede and hot-key concentration.
  5. •Monitor hit rate, latency, and eviction behavior to iterate policy.

Quality Gates

  • •Key naming and ownership are consistent across services.
  • •Invalidation strategy matches freshness requirements.
  • •Cache failures do not silently corrupt correctness expectations.
  • •Metrics allow quick detection of cache regressions.

Failure Handling

  • •Stop when invalidation policy cannot guarantee correctness bounds.
  • •Escalate when cache behavior increases tail latency or error rates.