← Back to Skills Marketplace
dylanbaker24

Memory Hygiene

by dylanbaker24 · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
25257
Downloads
19
Stars
249
Active Installs
1
Versions
Install in OpenClaw
/install memory-hygiene
Description
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.
Usage Guidance
Install only if you intentionally want a skill that can guide full Clawdbot vector-memory deletion. Before using the wipe or cron instructions, export or back up memory, confirm the exact path, keep a rollback plan, and avoid enabling monthly automation unless you are comfortable with recurring unattended memory resets.
Capability Analysis
Type: OpenClaw Skill Name: memory-hygiene Version: 1.0.0 The skill includes instructions for the AI agent to execute high-risk system commands. Specifically, `rm -rf ~/.clawdbot/memory/lancedb/` is used to wipe memory, and `cron action=add` is used to schedule persistent tasks. While these actions are presented for a benign 'memory hygiene' purpose, the direct execution of `rm -rf` and the ability to add cron jobs represent powerful capabilities that could be easily misused if the skill were compromised or the agent tricked.
Capability Assessment
Purpose & Capability
The audit, wipe, reseed, and auto-capture configuration guidance fits the stated memory-hygiene purpose, but the capability includes full deletion of the LanceDB vector-memory store.
Instruction Scope
SKILL.md instructs `rm -rf ~/.clawdbot/memory/lancedb/` to clear all vector memory and does not include an explicit backup, confirmation, dry-run, path validation, or recovery workflow.
Install Mechanism
The package is instruction-only with a single non-executable SKILL.md file and no install scripts or bundled executable code.
Credentials
The file paths and commands are scoped to Clawdbot memory maintenance, but the skill works with persistent memory and suggests storing sensitive-adjacent facts such as accounts, credential locations, and contacts while warning not to store OAuth URLs or tokens.
Persistence & Privilege
The monthly maintenance example adds a persistent cron job that repeatedly wipes and reseeds memory, with no per-run approval, disable instructions, rollback guidance, or verification that reseeding succeeded.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install memory-hygiene
  3. After installation, invoke the skill by name or use /memory-hygiene
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Version 1.0.0 – Initial Release - Introduces memory-hygiene for auditing, cleaning, and optimizing Clawdbot's vector memory (LanceDB). - Provides quick commands for memory audit, wipe, and reseed. - Details configuration for disabling auto-capture to prevent junk memories. - Lists guidelines on what types of information to store or avoid storing in memory. - Includes instructions for setting up monthly automated maintenance.
Metadata
Slug memory-hygiene
Version 1.0.0
License
All-time Installs 841
Active Installs 249
Total Versions 1
Frequently Asked Questions

What is 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. It is an AI Agent Skill for Claude Code / OpenClaw, with 25257 downloads so far.

How do I install Memory Hygiene?

Run "/install memory-hygiene" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Memory Hygiene free?

Yes, Memory Hygiene is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Memory Hygiene support?

Memory Hygiene is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Memory Hygiene?

It is built and maintained by dylanbaker24 (@dylanbaker24); the current version is v1.0.0.

💬 Comments