skills.rest
mainion-ai
mainion-aiCOMMUNITY
Explore all skills by this author →
GitHub

mk-memory-setup

Configure host-based memory-kernel environment for NanoClaw agent persistence.

12|1|Updated Mar 9, 2026
Search Tags:#setup#claude.md#cron#persistent-memory#nanoclaw#memory-kernel
One-click install
npx skills add https://github.com/mainion-ai/memory-kernel --skill mk-memory-setup
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mk-memory-setup
Source: https://github.com/mainion-ai/memory-kernel/tree/main/container/skills/mk-memory-setup
Command: npx skills add https://github.com/mainion-ai/memory-kernel --skill mk-memory-setup

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Memory data is typically lost between sessions, forcing agents to relearn and repeat decisions. This skill provides a host-level setup to install and configure memory-kernel, render CLAUDE.md for the agent, and maintain persistent storage across sessions.

Core Features & Use Cases

  • Install the memory-kernel CLI and initialize a persistent memory directory structure (ENTITIES, EPISODES, etc.).
  • Configure NanoClaw container mounts and mount allowlist to expose host memory to the agent.
  • Render CLAUDE.md so the agent loads its memory view at boot.
  • Optionally back up memory data to GitHub for long-term storage and collaboration.
  • Provide step-by-step post-setup tasks (cron nightly sync, restart, and verification).

Quick Start

Run mk-memory-setup to configure host memory for a NanoClaw agent, install memory-kernel, initialize memory directories, and render the initial CLAUDE.md.

Frequently Asked Questions about mk-memory-setup

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I set up persistent memory for a NanoClaw agent across sessions?▼

Persistent memory for a NanoClaw agent is set up by installing the memory-kernel CLI and initializing a host-based directory structure. This environment ensures agent memory data is maintained across sessions instead of being lost.

What dependencies do I need to configure host memory for an agent?▼

Configuring host memory requires Node 20+, Git, sqlite3, and Docker for NanoClaw. The setup validates these dependencies and optionally uses the GitHub CLI if you want to back up memory data.

How does a memory-kernel environment work with Docker containers?▼

A memory-kernel environment works with Docker by configuring NanoClaw container mounts and a mount allowlist. This exposes the host-level persistent storage directories to the agent running inside the container.

Can I back up agent memory data to GitHub automatically?▼

You can automatically back up agent memory data to GitHub by configuring a nightly cron sync job. This requires the optional GitHub CLI dependency to handle the long-term storage and collaboration push.

Why does my agent lose its memory context between sessions?▼

Agents lose memory context between sessions because memory data is typically ephemeral. Rendering a CLAUDE.md file at boot loads a persistent memory view, allowing the agent to recall past decisions without relearning them.

Related Skills

Explore More Software Engineering Skills→
flashinfer-aiflashinfer-ai
264

clone-repos

Clone and update SGLang and FlashInfer repositories with submodules from GitHub.

Official
Advanced
thmtzthmtz
5

manage-mounts

Configure and control host directory access for NanoClaw agent containers.

Community
Intermediate
wrightchinwrightchin

manage-mounts

Manage NanoClaw agent container access to host directories via a JSON allowlist.

Community
Intermediate
jonnymuirjonnymuir
1

nap

Compress agent history files and prune stale logs in the .squad/ directory.

Community
Basic