Topic
#disk-cleanup
Loot, blog posts and adjacent themes connected to this topic. Follow the tag to keep it in your orbit.
1Shown loot
0Shown articles
8Linked neighbor tags
Topic paths
If you want to go deeper, the adjacent tags are the fastest way to compare and branch into related workflows.
Loot
More from this topic
Keep OpenClaw's disk and memory from quietly filling up with stale files and duplicate notes
No votes yet
Text: AI-generated
AI-generated · Automatically published by LinkLoot. active-maintenance is an OpenClaw skill that checks disk usage, cleans aged temporary files, deduplicates memory fragments, and logs every maintenance decision for later review. AI-generated: This Loot was created and published automatically by LinkLoot and was not substantively reviewed by a human editor. What it does active-maintenance is a Python-based OpenClaw skill that runs periodic housekeeping on the agent's own environment. Four documented features: System health checks — monitors disk usage against a configurable percentage threshold and warns when resources get tight. Auto-cleanup — deletes temporary files and artifacts older than a configurable age (in days) from a defined list of directories (TEMPDIRS). Memory metabolism (M3) — removes exact duplicates from stored memory fragments and distills dense notes into shorter core insights, so the agent's long-term context stays useful instead of bloated. Decision logging — writes a record of each maintenance cycle to MEMORY/DECISIONS/, giving you an audit trail of what was cleaned or summarized and when. The skill ships as nightlyoptimizer.py plus a small decisionlogger module. A full run is one command: python3 /root/.openclaw/workspace/scripts/nightlyoptimizer.py. Configuration lives at the top of that script — the directory list, the disk-usage warning threshold, and the file-age cutoff. Who should use it OpenClaw users whose workspaces accumulate temporary files, build artifacts, or repeated memory entries over time. It is especially relevant for agents running scheduled or long-lived tasks where disk pressure and noisy memory degrade behavior gradually rather than all at once. Setup surface Requires a working OpenClaw workspace with Python 3 available. The documented invocation path assumes the script sits under /root/.openclaw/workspace/scripts/; adjust paths if your workspace differs. All tunables are plain variables in the script header — no external service, API key, or network access is described in the documentation. Runner test plan No execution testing has been performed. A proposed static check would verify the script parses, imports resolve, and configuration defaults are sane. Dependency review covers only standard-library-style assumptions visible in the docs. An injection pass would confirm that logged decision titles cannot write outside MEMORY/DECISIONS/. A sandboxed dry run would inspect which paths cleanup actually touches before any deletion occurs. Screenshots of a sample decision log would help confirm output format. Risk notes Auto-cleanup deletes files by age; a misconfigured TEMPDIRS entry or too-short days value could remove data you wanted kept. Memory distillation rewrites notes, so original wording may be lost — keep backups of the memory directory before first run. The skill is authored for a specific assistant setup ("Kim Assistant") and inspired by another project; portability beyond that layout is unverified. Pricing and licensing terms were not included in the supplied packet, so they remain unknown. Source links Original skill documentation Skill listing Awesome OpenClaw Skills ClawHub registry
- LinkLoot access
- Free
- Provider costs
- Unknown
Review open
Blog
Related reads
No blog posts for #disk-cleanup yet
There is no published article with this tag right now. Browse the blog for adjacent themes or follow the tag for future updates.