ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified developer tools Safety 2/5

memory-hygiene

Audit, clean, and optimize Clawdbot's vector memory (LanceDB). Use when memory is bloated with junk, token usage is high from irrelevant auto-recalls, or setting up memory maintenance automation.

Why use this skill?

Clean up and optimize your Clawdbot vector memory with the Memory Hygiene skill. Prevent token waste, prune junk data, and automate maintenance.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/dylanbaker24/memory-hygiene
Or

What This Skill Does

Memory Hygiene is a specialized OpenClaw skill designed to maintain the integrity and efficiency of the Clawdbot vector memory system (LanceDB). Over time, automated memory capture can lead to "memory bloat," where irrelevant data, transient system status messages, and repeated heartbeat signals consume expensive token windows and degrade the relevance of AI context retrieval. This skill provides the necessary tooling to audit, selectively prune, and architect a cleaner long-term memory strategy for your agent.

Installation

To install this skill, use the Clawhub CLI within your terminal: clawhub install openclaw/skills/skills/dylanbaker24/memory-hygiene

Use Cases

  • Automated Memory Cleanup: Configure recurring tasks to prevent long-term token degradation.
  • Junk Filtering: Stop the automated intake of transient data like heartbeat logs and OAuth redirects.
  • Precision Recall: Ensure that retrieved memories consist of high-value project facts, user preferences, and architectural decisions rather than noise.
  • System Recovery: Perform a "fresh start" on the vector database after a project lifecycle concludes.

Example Prompts

  1. "Perform an audit of my current memory storage by listing the last 50 entries to see if we have significant junk accumulation."
  2. "Update my memory configuration to disable auto-capture, as I want to curate my vector database manually to save on token costs."
  3. "Schedule a monthly maintenance task that wipes and reseeds our core knowledge base using the facts in MEMORY.md."

Tips & Limitations

  • Conciseness Matters: Aim for entries under 100 words per memory to maximize the efficacy of vector search retrieval.
  • Strategic Storage: Use the importance parameter (0.7-1.0) to prioritize critical information. Entries with lower importance scores are the best candidates for pruning during manual audits.
  • Safety Note: Be extremely careful when using rm -rf on your database path. Ensure you have backed up any critical persistent data stored in ~/.clawdbot/memory/lancedb/ before triggering a full wipe. This tool is designed for developers who maintain a separate, human-readable record of critical facts in a MEMORY.md file, which should serve as the source of truth for reseeding.

Metadata

Stars2387
Views6
Updated2026-03-09
View Author Profile
AI Skill Finder

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 skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-dylanbaker24-memory-hygiene": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#memory-management#vector-database#token-optimization#system-maintenance#clawdbot
Safety Score: 2/5

Flags: file-write, file-read