Cache Keepalive
OpenCode plugin that uses bounded provider-request replay to keep prompt caches warm, with a live TUI readout
0
895
近 7 天 616
38.1
生态多维模型
4 天前
2026-09-30
快速安装与配置
opencode.json写入当前项目的 opencode.json,只对这个仓库生效。
opencode.json
{
"$schema": "https://opencode.ai/config.json",
"plugin": ["@vikrant82/opencode-cache-keepalive@0.2.0"]
}写入 ~/.config/opencode/opencode.json,对所有项目生效。
~/.config/opencode/opencode.json
{
"$schema": "https://opencode.ai/config.json",
"plugin": ["@vikrant82/opencode-cache-keepalive@0.2.0"]
}若你要在本地改造这个插件,先装到项目里再从本地路径引用。
shell
pnpm add -D @vikrant82/opencode-cache-keepaliveOpenCode 启动时会通过内嵌运行时自动加载 npm 依赖并缓存至本地目录,无需手动在全局环境执行安装。
该插件暂未提供 README 文档
缺少文档通常代表该项目仍在开发初期。建议前往其源码仓库查看详细实现。
同类生态推荐
Kiro Provider
opencode-kiro-provider
OpenCode provider for Kiro: every model your Kiro subscription offers (GPT-5.6, Claude Opus 5, Sonnet, open-weight), live model list, reuses your kiro-cli login
Llm Proxy
opencode-llm-proxy
Local AI gateway for OpenCode with tool/function calling — use any model via OpenAI, Anthropic, or Gemini API format
Magic Compact
magic-compact
Lossless context compression plugin for OpenCode.