Token导航 LogoToken导航TokenDH.com
研究检索可写文件clawhub未标认证来源可访问clear审计通过

save-load节省负载

Agent Skill

save-load 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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

3,794

周安装

155

GitHub Stars

公开资料未说明

下载量

1,215
OpenClaw

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

MIT-0

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:save-load(节省负载)
来源仓库:https://github.com/jkyotnfjfbjnknh/save-load
安装命令:
openclaw skills install save-load
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install save-load

简介

save-load 提供 /save 和 /load 斜杠命令保存与加载 OpenClaw 对话上下文。

  • 适合需要持久化会话状态、恢复中断任务或多轮调试的场景。
  • 通过 clawhub 安装后,Agent 可在交互中调用保存与加载指令。
  • 使用前需确认存储后端可用性及数据隐私保护机制。
  • save-load 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

name
save-load
description
Save and load OpenClaw conversation context with /save and /load slash commands
metadata

Save/Load Context Plugin

An OpenClaw plugin that adds /save and /load slash commands for persisting and restoring conversation context across sessions.

Problem

OpenClaw conversations are ephemeral — when you start a new session or reset, all context is lost. This plugin lets you save any conversation to disk and reload it later, preserving full context continuity.

Commands

CommandDescription
/save [label]Save the current conversation context with an optional label. If no label is provided, the save is named "unnamed".
/loadList all previously saved conversation contexts with their labels, message counts, and save dates.
/load <name>Load a saved context by label or index number. This replaces the current session transcript with the saved conversation. After loading, the agent receives a system event with the save file path so it can restore full context.
/load --delete <name>Permanently delete a saved context file.

How It Works

Saving

  1. When /save is invoked, the plugin reads the session transcript (JSONL format)
  2. It identifies the conversation boundary — messages since the last /new, /reset, or /load
  3. Only user and assistant messages are extracted (tool calls and results are excluded)
  4. The messages are written to ~/.openclaw/saves/<timestamp>_<label>.json
  5. An index file (index.json) is updated for fast listing

Loading

  1. /load <name> finds the save file by label or index
  2. The saved messages are written as a new session transcript
  3. A system event is injected via enqueueSystemEvent() containing the save file path and message count
  4. The agent receives this event and can use the read tool to load the full context from the save file

Save File Format

{
  "version": 1,
  "label": "project-discussion",
  "savedAt": "2026-03-30T15:38:00+08:00",
  "messageCount": 42,
  "sourceSession": "telegram:2074807638",
  "messages": [
    { "role": "user", "content": "..." },
    { "role": "assistant", "content": "..." }
  ]
}

Installation

Via OpenClaw Plugin Manager

openclaw plugins install save-load

Manual Installation

  1. Copy the plugin directory to ~/.openclaw/extensions/save-load/
  2. Ensure openclaw.plugin.json and package.json are present
  3. Add "save-load" to the plugins.enabled array in ~/.openclaw/openclaw.json
  4. Restart the OpenClaw gateway: openclaw gateway restart

Storage

Save files are stored in ~/.openclaw/saves/. There is no limit on the number or size of saves — only disk space applies.

Requirements

  • OpenClaw 2026.3.30 or later
  • Node.js (included with OpenClaw)

Source Code

https://github.com/jkyotnfjfbjnknh/openclaw-save-load

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenClaw

83.91%
按下载量换算1,020

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

可写文件

该 Skill 可能写入或修改本地文件,使用前需要确认目标目录和修改范围。

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