context-cleanup
Analyze and archive low-value memory notes in OpenClaw workspace to reduce context bloat and improve responsiveness. 适用于上下文冗余、维护整理场景;默认流程为 analyze → plan → confirm → archive,归档优先于删除。
Why use this skill?
Optimize your OpenClaw workspace performance by automatically identifying and archiving low-value memory notes. Keep your context clean, lean, and highly responsive.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/irideas/context-cleanupWhat This Skill Does
The context-cleanup skill is a specialized utility for the OpenClaw agent designed to maintain a lean, efficient, and responsive workspace. As AI agents interact with long-term projects, the accumulated memory logs often become cluttered with outdated, trivial, or redundant information, commonly referred to as 'context bloat.' This skill proactively identifies low-value memory notes, organizes them, and moves them to archival storage instead of permanently deleting them, ensuring your workspace remains performant without sacrificing historical data.
Installation
To install this skill, run the following command in your terminal within the OpenClaw environment:
clawhub install openclaw/skills/skills/irideas/context-cleanup
Use Cases
- Project Housekeeping: Regularly clearing out temporary design thoughts or abandoned branches from memory to keep the model's focus on active tasks.
- Context Optimization: Reducing the token count for long-running sessions, which directly improves the speed and accuracy of agent responses.
- Workspace Auditing: Reviewing what the agent deems 'low-value' to get a better understanding of how the model categorizes information.
- Compliance & Archiving: Safely storing older context in a structured way that avoids deletion but keeps it out of the active working set.
Example Prompts
- 'Hey OpenClaw, please run a context cleanup analysis to see how much bloat is in our current workspace.'
- 'I need to free up memory; can you generate an archival plan for all notes older than 2026-03-01?'
- 'Start the cleanup process, but please show me the plan first and wait for my approval before archiving anything.'
Tips & Limitations
- Always verify the plan: While the skill includes a
--dry-runflag, you should always review theplanoutput before proceeding with thearchivestep to ensure important documents aren't inadvertently moved. - Safety Boundaries: The skill is designed to never touch protected files such as
MEMORY.md,specs/, orAGENTS.md. These files are considered critical for agent function and are excluded by default. - Automation: Use the
--yesflag with caution. Only enable it in non-interactive CI/CD pipelines where you have established robust pre-check routines. - Manual Recovery: Since the skill uses an archive method rather than a hard delete, you can always recover files if the agent misclassifies an item, provided you know where the archive directory is stored.
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-irideas-context-cleanup": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-read, file-write, code-execution