← Back to Skills Marketplace
sopaco

memclaw-maintance

by Sopaco · GitHub ↗ · v0.9.31 · MIT-0
cross-platform ✓ Security Clean
228
Downloads
1
Stars
0
Active Installs
7
Versions
Install in OpenClaw
/install memclaw-maintance
Description
MemClaw Maintenance Guide — Installation, configuration, and maintenance guidance. For daily usage and tool operations, use the [`memclaw` skill](https://cla...
Usage Guidance
This is an instruction-only maintenance guide and appears internally consistent, but before proceeding: 1) Verify the authenticity of the @memclaw/memclaw plugin and the referenced GitHub repo (review its code and releases) because the plugin installation will fetch and run code on your machine. 2) Backup your existing OpenClaw workspace (~/.openclaw/workspace) before running migration; migration reads and converts local memory files. 3) Be deliberate when editing AGENTS.md/openclaw.json: the guide recommends making MemClaw the default memory path and always loading persistent memory at conversation start, which changes agent behavior. 4) Ensure the LLM and embedding API keys you enter are for the intended providers and keep them secret; the guide says keys are stored as sensitive but confirm your OpenClaw config practices. 5) Check port usage (6333, 6334, 8085) and firewall rules if the plugin auto-starts services. If you want to reduce risk, inspect the plugin package source before installing and test on a non-production environment first.
Capability Analysis
Type: OpenClaw Skill Name: memclaw-maintance Version: 0.9.31 The skill bundle is a legitimate maintenance and installation guide for the MemClaw memory plugin (associated with the sopaco/cortex-mem GitHub project). It provides instructions for the agent to assist users with plugin installation, API key configuration in openclaw.json, and updating agent guidelines (AGENTS.md) to integrate new memory tools like cortex_search and cortex_migrate. While it involves high-privilege actions such as modifying configuration files and handling API keys, these behaviors are transparently documented and strictly aligned with the stated purpose of system maintenance.
Capability Assessment
Purpose & Capability
Name and description match the content: the skill is an installation/configuration/maintenance guide for MemClaw. All actions described (install plugin, enable plugin, configure LLM/embedding keys, migrate local memories, run maintenance commands) are coherent with a maintenance role. Nothing in the files asks for unrelated cloud provider credentials or unrelated system-wide access.
Instruction Scope
The SKILL.md instructs the agent/admin to edit openclaw.json / AGENTS.md and to run migration that reads local memory files (~/.openclaw/workspace/memory/) and to call MemClaw runtime endpoints (cortex_* calls). Those actions are expected for migration/maintenance but are capable of reading local memory files and changing agent behavior (the guide insists on making the agent load persistent memory at conversation start). Users should be aware migration reads local files and that modifying AGENTS.md affects agent startup behavior.
Install Mechanism
This skill contains no install spec or code — it is instruction-only. It tells the user to run the OpenClaw CLI to install @memclaw/memclaw, and points to a GitHub repository. The skill itself does not automatically download or execute code, so the installation risk is delegated to the plugin installation step (not part of this skill bundle).
Credentials
The skill does not declare environment variables, but instructs configuring LLM and embedding API keys in OpenClaw plugin settings (llmApiKey, embeddingApiKey, base URLs, models). These are proportionate to a memory/embedding plugin, but users should confirm API keys are only for their intended LLM/embedding providers and understand they will be stored in OpenClaw config (marked sensitive).
Persistence & Privilege
The skill does not request 'always: true' or special privileges. It does instruct persistent changes to agent configuration (AGENTS.md / openclaw.json) so the MemClaw behavior becomes a default; that is expected for a maintenance guide but is effectively a persistent change to agent behavior and should be reviewed before applying.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install memclaw-maintance
  3. After installation, invoke the skill by name or use /memclaw-maintance
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.9.31
- Adds a new required step instructing users to update their AGENTS.md guideline for full MemClaw compatibility. - Provides detailed markdown to replace the legacy "Memory" section in AGENTS.md, ensuring future sessions use MemClaw instead of older methods. - Warns that skipping this step may cause agents to ignore MemClaw and continue using legacy memory. - No code or file changes detected—documentation only.
v0.9.26
Version 0.9.27 - No file changes detected in this release. - Documentation, installation guidance, and maintenance instructions remain unchanged. - All guidance, security information, and references in SKILL.md are the same as the previous version.
v0.9.25
Version 0.9.23 - Updated the Security & Trust section to remove mention of data not being sent to external servers and clarify that API keys are not transmitted except to configured providers. - No code or functional changes; documentation update only.
v0.9.22
- Updated MemClaw overview to highlight improved search precision, lower token consumption, and enhanced memory retention over built-in memory. - Removed references to service startup and API key handling from the maintenance guide, focusing documentation on local storage and migration. - Added a note that both MemClaw and this skill are open source, with GitHub links. - Clarified that all processing is local and API keys are not transmitted except to LLM/embedding providers. - No functional or file changes in this release; documentation enhancements only.
v0.9.21
- Updated setup instructions to include disabling memorySearch in agent defaults during plugin enablement. - Clarified the need to set "memorySearch" to false in the "agents.defaults" section of openclaw.json during installation. - No changes to functionality—documentation only.
v0.9.20
- Added instructions to verify required configuration fields for LLM and Embedding during setup. - Guidance now includes proactively prompting users to complete required plugin configuration if items are missing. - Clarified that OpenClaw Gateway must be restarted after changes for configuration to take effect. - No other changes to installation, maintenance, or tool operation instructions.
v0.9.18
- Initial release of the memclaw-maintance skill. - Provides installation, configuration, and maintenance guidance for the MemClaw memory system. - Clarifies the distinction between maintenance (this skill) and daily usage (use the separate memclaw skill). - Includes setup steps, security details, troubleshooting resources, and guidance on migration and maintenance operations. - Details data storage locations and confirms all operations and data remain local.
Metadata
Slug memclaw-maintance
Version 0.9.31
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 7
Frequently Asked Questions

What is memclaw-maintance?

MemClaw Maintenance Guide — Installation, configuration, and maintenance guidance. For daily usage and tool operations, use the [`memclaw` skill](https://cla... It is an AI Agent Skill for Claude Code / OpenClaw, with 228 downloads so far.

How do I install memclaw-maintance?

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

Is memclaw-maintance free?

Yes, memclaw-maintance is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does memclaw-maintance support?

memclaw-maintance is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created memclaw-maintance?

It is built and maintained by Sopaco (@sopaco); the current version is v0.9.31.

💬 Comments