菜单
Apply when implementing caching, session storage, rate limiting, or pub/sub with Redis. Covers: key design, TTL strategy, cache invalidation, Redis Streams, connection pooling. Trigger for: Redis, cache, session, rate limit, pub/sub, queue.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/Tryboy869/dojutsu-for-ai --skill redis-caching命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Apply for abp-net-development-rules. You are a senior .NET backend developer and an expert in C#, ASP.NET Core, ABP Framework, and Entity Framework Core. - Write concise, idiomatic C# code with accurate examples. - Follow ABP Framework’s
[Applies to: **/*.rs] Definitive guidelines for building high-performance, maintainable, and secure web applications with actix-web 4, focusing on modern Rust best practices.
Apply for additional-resources. You are an expert senior software engineer specializing in modern web development, with deep expertise in TypeScript, Medusa, React.js, and TailwindCSS. - Don't use type aliases when importing files.
基于 SOC 职业分类
正在显示 SKILL.md
| name | redis-caching |
| description | Apply when implementing caching, session storage, rate limiting, or pub/sub with Redis. Covers: key design, TTL strategy, cache invalidation, Redis Streams, connection pooling. Trigger for: Redis, cache, session, rate limit, pub/sub, queue. |
# Always namespace to avoid collisions
f"session:{user_id}" # user session
f"cache:user:{user_id}" # cached user data
f"rate:{endpoint}:{ip}" # rate limit counter
f"lock:job:{job_id}" # distributed lock
f"worker:{worker_id}:alive" # heartbeat
# Key limits: < 512 bytes, avoid spaces
import redis.asyncio as redis
r = redis.from_url("redis://localhost:6379/0")
# Always set TTL — never store without expiry
await r.setex("cache:user:123", 300, json.dumps(user_data)) # 5 min
# Check before compute (cache-aside pattern)
async def get_user(user_id: str) -> dict:
cached = await r.get(f"cache:user:{user_id}")
if cached:
return json.loads(cached)
user = await db.fetch_user(user_id)
await r.setex(f"cache:user:{user_id}", 300, json.dumps(user))
return user
async def with_lock(key: str, ttl: int = 30):
"""Distributed lock using SET NX EX."""
lock_key = f"lock:{key}"
acquired = await r.set(lock_key, "1", nx=True, ex=ttl)
if not acquired:
raise LockError(f"Could not acquire lock: {key}")
try:
yield
finally:
await r.delete(lock_key)
async def is_rate_limited(identifier: str, limit: int, window: int) -> bool:
key = f"rate:{identifier}"
pipe = r.pipeline()
pipe.incr(key)
pipe.expire(key, window)
results = await pipe.execute()
return results[0] > limit
pool = redis.ConnectionPool.from_url(
"redis://localhost:6379/0",
max_connections=50,
decode_responses=True,
)
r = redis.Redis(connection_pool=pool)
❌ Keys without TTL (Redis will fill memory)
❌ KEYS * in production (blocks Redis, use SCAN)
❌ Storing large objects (> 1MB) — use S3/filesystem
❌ Using Redis as primary database (it's ephemeral)
❌ No connection pooling
❌ Storing sensitive data without encryption