skills.rest
chiptoe-svg
chiptoe-svgCOMMUNITY
Explore all skills by this author →
GitHub

add-mnemon

Add persistent graph-based memory to NanoClaw agent containers.

5|Updated Mar 14, 2026
Search Tags:#ai agents#persistent memory#nanoclaw#context recall#container configuration#mnemon#graph memory
One-click install
npx skills add https://github.com/chiptoe-svg/nanoclaw_gccourse --skill add-mnemon-chiptoe-svg
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: add-mnemon
Source: https://github.com/chiptoe-svg/nanoclaw_gccourse/tree/main/.claude/skills/add-mnemon
Command: npx skills add https://github.com/chiptoe-svg/nanoclaw_gccourse --skill add-mnemon-chiptoe-svg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI agents running in isolated containers lose all conversational context and learned insights when the container restarts, forcing users to restate background information and preventing the agent from building on past interactions.

Core Features & Use Cases

  • Persistent Graph-Based Memory: Installs the mnemon tool in the agent container to store insights and context in a structured graph format.
  • Automatic Context Injection: For Claude Code providers, registers hooks to surface relevant past context before the agent responds and store new insights after each turn; for OpenCode providers, injects memory context directly into every prompt.
  • Cross-Session Continuity: Memory is stored in a per-agent-group mounted directory, so it survives container restarts and is accessible across all agent sessions.
  • Use Case: A classroom AI assistant can remember student questions from previous office hours, reference past lesson plans, and build on earlier troubleshooting sessions without the instructor repeating context.

Quick Start

Use the add-mnemon skill to add persistent graph-based memory to your NanoClaw agent container, configured for your active provider to recall past context and store insights across restarts.

Frequently Asked Questions about add-mnemon

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

FAQPage Schema
How do I give AI agents persistent memory across container restarts?▼

You can give AI agents persistent memory across container restarts by installing a graph-based memory tool that stores context in a per-agent-group mounted directory, ensuring data survives the container lifecycle. The add-mnemon skill automates this setup for NanoClaw deployments.

How does graph-based context recall work for long-running agent systems?▼

Graph-based context recall works by storing conversational insights in a structured graph format within a mounted directory. The agent retrieves relevant past context before responding and saves new insights after each interaction, maintaining continuity across sessions.

Does persistent cross-session memory work with both Claude Code and OpenCode providers?▼

Yes, persistent cross-session memory works with both providers. For Claude Code, it uses hook-based context injection to surface past context and store insights. For OpenCode, it injects memory context directly into every prompt to maintain continuity.

How do I stop my classroom AI assistant from forgetting past lesson plans and student questions?▼

To stop a classroom AI assistant from forgetting past lessons, apply persistent cross-session memory using a graph structure. This allows the agent to reference previous office hours questions and earlier troubleshooting sessions without the instructor repeating context.

What are the limitations of using mounted directories for agent context recall?▼

The primary limitation of using mounted directories for agent context recall is that memory is scoped per-agent-group. While it survives container restarts, the stored graph memory is tied to the specific NanoClaw deployment configuration and requires a compatible provider like Claude Code or OpenCode.

Related Skills

Explore More Software Engineering Skills→
chiptoe-svgchiptoe-svg

add-opencode

Integrate the OpenCode agent provider into NanoClaw for alternative LLM backends.

Community
Intermediate
salmanparachasalmanparacha

memory-systems

Store, retrieve, and manage persistent agent memories across sessions.

Community
Advanced
Hex-4Hex-4

hook-development

Create and manage Claude Code plugin hooks for event-driven workflow automation.

Community
Advanced
zilliztechzilliztech
3

chat-memory

Store and retrieve past chatbot conversations across sessions using vector memory.

Official
Advanced