context-continuity
上下文续接与记忆管理。当用户说"新建对话"、"继续"、"接着上次"等时, 自动从 MEMORY.md 和 memory/ 文件中提取当前任务/项目上下文, 让新会话能够了解之前的进展。用于:(1) 用户要开始新对话但想保留上下文 (2) 用户说"继续上次的工作" (3) 用户想要回顾当前进行中的任务 特点:自动记录所有对话内容到 memory/YYYY-MM-DD.md,无需询问用户
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/cjstate/context-continuityWhat This Skill Does
The Context Continuity skill is a robust memory management system designed for OpenClaw. It automatically persists conversation history into structured Markdown files located in the memory/ directory. By tracking ongoing projects, user decisions, and pending tasks without requiring explicit user commands to save, it ensures that your AI assistant remains state-aware across different sessions. When triggered, the skill scans your daily logs to synthesize a coherent summary of your current progress, allowing for a seamless transition between work sessions. It effectively maintains both short-term daily logs and long-term project insights via MEMORY.md.
Installation
To integrate this skill into your OpenClaw environment, execute the following command in your terminal:
clawhub install openclaw/skills/skills/cjstate/context-continuity
Ensure that your environment has write permissions to the ~/.openclaw/workspace/ directory, as the skill needs to create and update log files automatically as part of its core functionality.
Use Cases
- Project Handoffs: Pick up where you left off after a break or overnight, as the AI remembers your specific technical decisions or draft progress.
- Task Tracking: Keep a running log of completed versus pending items without manually maintaining a separate document.
- Contextual Recall: When switching tasks frequently, use the skill to remind the AI about the constraints or requirements established in previous sessions.
- Automated Documentation: Since all dialogues are logged to
memory/YYYY-MM-DD.md, users can treat these files as an automated journal of their problem-solving process.
Example Prompts
- "新建对话,我们要继续上次关于购物平台的开发工作。"
- "回顾一下我们之前讨论的技术选型,看看还有哪些待办事项没处理。"
- "接着上次的逻辑,帮我把这段代码补充完整。"
Tips & Limitations
- Privacy: Note that the skill stores content locally. If you handle sensitive data, ensure your local directory is encrypted or protected.
- Maintenance: While the skill handles recording, we recommend manually reviewing your
memory/folder periodically. The skill does not prune files automatically; consider archiving logs older than 30 days to keep your workspace tidy. - Conciseness: The skill works best when you provide clear, structured feedback during the conversation, as it generates summaries based on your input. Avoid ambiguous phrasing if you want highly accurate recall later.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-cjstate-context-continuity": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-write, file-read