Token导航 LogoToken导航TokenDH.com
前端设计只读github未标认证来源可访问许可证需确认审计通过

go-cache去缓存

Agent Skill

go-cache 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

456

周安装

19

GitHub Stars

公开资料未说明

下载量

152
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:go-cache(去缓存)
来源仓库:https://github.com/cristiano-pacheco/ai-tools
仓库路径:skills/go-cache
安装命令:
npx skills add https://github.com/cristiano-pacheco/ai-tools --skill go-cache
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/cristiano-pacheco/ai-tools --skill go-cache

简介

用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态进行整理。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装,需确认权限范围。
  • 使用前建议核验是否会触发联网或文件读写操作。
  • go-cache 属于前端设计类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Go Cache

Generate two files for every cache: a port interface and a Redis-backed implementation.

When to Use

  • Create a cache layer for any module
  • Redis-backed TTL storage (OTP, sessions, OAuth state)
  • Rate limiting storage
  • Boolean flag caching (existence checks)
  • JSON data caching (structured objects)

Which Variant?

Pick before writing anything:

ScenarioVariantGet return type
Flag, existence check, rate limitBoolean flagbool
Structured data — tokens, sessions, profilesJSON data*dto.XxxData

For TTL:

  • Fixed TTL — short-lived or individually written entries (OTPs, OAuth state, rate limits, sessions)
  • Randomized TTL — long-lived entries written in bulk (activation flags, daily metrics) — prevents cache stampede

Two-File Pattern

Every cache requires exactly two files:

  1. Port interface: internal/modules/<module>/ports/<cache_name>_cache.go
  2. Cache implementation: internal/modules/<module>/cache/<cache_name>_cache.go

File Layout Order

  1. Constants (key prefix, TTL)
  2. Implementation struct (XxxCache)
  3. Compile-time interface assertion
  4. Constructor (NewXxxCache)
  5. Methods (Set, Get, Delete)
  6. Helper methods (buildKey, calculateTTL)

Boolean Flag Cache

Use when caching simple existence flags, presence checks, or rate limit states.

  • Store "1" as the value
  • Return false, nil when the key doesn't exist (not an error)

Port

package ports

import "context"

// XxxCache describes ...
type XxxCache interface {
	Set(ctx context.Context, id uint64) error
	Get(ctx context.Context, id uint64) (bool, error)
	Delete(ctx context.Context, id uint64) error
}

Implementation

package cache

import (
	"context"
	"errors"
	"fmt"
	"time"

	"github.com/cristiano-pacheco/bricks/pkg/redis"
	"github.com/cristiano-pacheco/pingo/internal/modules/<module>/ports"
	redislib "github.com/redis/go-redis/v9"
)

const (
	entityCacheKeyPrefix = "entity_name:"
	entityCacheTTL       = 10 * time.Minute
)

type EntityCache struct {
	redisClient redis.UniversalClient
}

var _ ports.EntityCache = (*EntityCache)(nil)

func NewEntityCache(redisClient redis.UniversalClient) *EntityCache {
	return &EntityCache{
		redisClient: redisClient,
	}
}

func (c *EntityCache) Set(ctx context.Context, id uint64) error {
	key := c.buildKey(id)
	return c.redisClient.Set(ctx, key, "1", entityCacheTTL).Err()
}

func (c *EntityCache) Get(ctx context.Context, id uint64) (bool, error) {
	key := c.buildKey(id)
	result := c.redisClient.Get(ctx, key)
	if err := result.Err(); err != nil {
		if errors.Is(err, redislib.Nil) {
			return false, nil
		}
		return false, err
	}
	return true, nil
}

func (c *EntityCache) Delete(ctx context.Context, id uint64) error {
	key := c.buildKey(id)
	return c.redisClient.Del(ctx, key).Err()
}

func (c *EntityCache) buildKey(id uint64) string {
	return fmt.Sprintf("%s%d", entityCacheKeyPrefix, id)
}

JSON Data Cache

Use when caching structured data. Data structs are defined in the dto package, never in ports.

  • Serialize with json.Marshal before storing
  • Deserialize with json.Unmarshal when retrieving
  • Return nil, nil on missing key — unless the key is always expected to exist, in which case return a domain error (e.g., errs.ErrXxxNotFound)
  • Use distinct variable names (getErr, unmarshalErr) to avoid shadowing

Port

package ports

import (
	"context"

	"github.com/cristiano-pacheco/pingo/internal/modules/<module>/dto"
}

// XxxCache describes ...
type XxxCache interface {
	Set(ctx context.Context, key string, data dto.XxxData) error
	Get(ctx context.Context, key string) (dto.XxxData, error)
	Delete(ctx context.Context, key string) error
}

Implementation

package cache

import (
	"context"
	"encoding/json"
	"errors"
	"fmt"
	"time"

	"github.com/cristiano-pacheco/bricks/pkg/redis"
	"github.com/cristiano-pacheco/pingo/internal/modules/<module>/dto"
	"github.com/cristiano-pacheco/pingo/internal/modules/<module>/ports"
	redislib "github.com/redis/go-redis/v9"
)

const (
	entityCacheKeyPrefix = "entity_name:"
	entityCacheTTL       = 10 * time.Minute
)

type EntityCache struct {
	redisClient redis.UniversalClient
}

var _ ports.EntityCache = (*EntityCache)(nil)

func NewEntityCache(redisClient redis.UniversalClient) *EntityCache {
	return &EntityCache{
		redisClient: redisClient,
	}
}

func (c *EntityCache) Set(ctx context.Context, key string, data dto.EntityData) error {
	cacheKey := c.buildKey(key)
	jsonData, err := json.Marshal(data)

	if err != nil {
		return fmt.Errorf("marshal entity data: %w", err)
	}

	return c.redisClient.Set(ctx, cacheKey, jsonData, entityCacheTTL).Err()
}

func (c *EntityCache) Get(ctx context.Context, key string) (dto.EntityData, error) {
	cacheKey := c.buildKey(key)
	result := c.redisClient.Get(ctx, cacheKey)

	if getErr := result.Err(); getErr != nil {
		if errors.Is(getErr, redislib.Nil) {
			return dto.EntityData{}, nil
		}
		return dto.EntityData{}, getErr
	}

	jsonData, err := result.Bytes()

	if err != nil {
		return dto.EntityData{}, fmt.Errorf("get bytes: %w", err)
	}

	var entityData dto.EntityData
	if unmarshalErr := json.Unmarshal(jsonData, &entityData); unmarshalErr != nil {
		return dto.EntityData{}, fmt.Errorf("unmarshal entity data: %w", unmarshalErr)
	}

	return entityData, nil
}

func (c *EntityCache) Delete(ctx context.Context, key string) error {
	cacheKey := c.buildKey(key)
	return c.redisClient.Del(ctx, cacheKey).Err()
}

func (c *EntityCache) buildKey(key string) string {
	return entityCacheKeyPrefix + key
}

Key Building

String ID (simple concatenation):

func (c *EntityCache) buildKey(id string) string {
	return entityCacheKeyPrefix + id
}

Uint64 ID:

func (c *EntityCache) buildKey(id uint64) string {
	return fmt.Sprintf("%s%d", entityCacheKeyPrefix, id)
}

Composite key:

func (c *EntityCache) buildKey(userID uint64, resourceID string) string {
	return fmt.Sprintf("%s%d:%s", entityCacheKeyPrefix, userID, resourceID)
}

TTL Configuration

Fixed TTL — for short-lived data where stampede is not a concern:

const (
	entityCacheKeyPrefix = "entity_name:"
	entityCacheTTL       = 10 * time.Minute
)

Randomized TTL — for long-lived data created in bulk (prevents cache stampede):

import "math/rand"

const (
	entityCacheKeyPrefix = "entity_name:"
	entityCacheTTLMin    = 23 * time.Hour
	entityCacheTTLMax    = 25 * time.Hour
)

func (c *EntityCache) calculateTTL() time.Duration {
	min := entityCacheTTLMin.Milliseconds()
	max := entityCacheTTLMax.Milliseconds()
	randomMs := min + rand.Int63n(max-min+1)
	return time.Duration(randomMs) * time.Millisecond
}

Common TTL ranges:

  • 5-15 minutes — OTP codes, OAuth state, rate limits
  • 50-70 minutes — User sessions
  • 12-25 hours — Activation flags, daily metrics
  • 6.5-7.5 days — Weekly aggregations

Naming

  • Port interface: XxxCache (ports package, no suffix)
  • Implementation struct: XxxCache (cache package — same name, disambiguated by package)
  • Constructor: NewXxxCache, returns *XxxCache
  • Constants: lowercase, package-level (e.g. entityCacheKeyPrefix, entityCacheTTL)

Fx Wiring

Add to internal/modules/<module>/module.go:

fx.Provide(
	fx.Annotate(
		cache.NewXxxCache,
		fx.As(new(ports.XxxCache)),
	),
),

Dependencies

  • redis.UniversalClient from "github.com/cristiano-pacheco/bricks/pkg/redis"
  • redislib "github.com/redis/go-redis/v9" for nil detection

Critical Rules

  1. No standalone functions: When a file contains a struct with methods, do not add standalone functions. Use private methods on the struct instead.
  2. Two files: Port in ports/, implementation in cache/
  3. Interface assertion: var _ ports.XxxCache = (*XxxCache)(nil) immediately below the struct
  4. Constructor: Returns *XxxCache (pointer)
  5. Context: Always accept ctx context.Context as first parameter — never call context.Background() internally
  6. Redis nil: Import redislib "github.com/redis/go-redis/v9" and check with errors.Is(err, redislib.Nil)
  7. TTL scope: TTL is an implementation detail — never expose it as a method parameter
  8. buildKey: Always use a buildKey() helper; + for string IDs, fmt.Sprintf for numeric IDs
  9. Missing keys: Boolean cache returns false, nil; JSON cache returns nil, nil (or a domain error if the key must exist)
  10. DTOs in dto package: Data structs belong in dto/, never defined inline in ports/
  11. No method comments: Only port interfaces get doc comments; implementation methods do not
  12. Error messages: "action noun: %w" format (e.g., "marshal oauth state: %w", "get bytes: %w")

Workflow

  1. Decide variant: Boolean flag or JSON data?
  2. Create port interface in ports/<name>_cache.go
  3. Create cache implementation in cache/<name>_cache.go
  4. Add Fx wiring to module.go
  5. Run make lint
  6. Run make nilaway

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.82%
按下载量换算51

Claude

29.63%
按下载量换算45

Cursor

19.83%
按下载量换算30

Gemini CLI

9.24%
按下载量换算14

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills