Memory Manager for Openclaw

An intelligent memory management system that automates session recording, daily summaries, and weekly distillations to build persistent long-term AI memory.

sukimgit
v1.0.0
Mar 6, 2026
0
118
1

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install auto-memory-manager

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install auto-memory-manager using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Memory Manager?

Memory Manager is a specialized utility designed to give AI agents a genuine long-term memory. Instead of losing context between sessions, this tool systematically captures interactions, extracts core decisions, and organizes them into a hierarchical memory structure. It acts as the cognitive backbone for developers using Openclaw Skills, ensuring that every project milestone and technical decision is preserved and retrievable.

The system bridges the gap between ephemeral chat history and permanent knowledge bases. By categorizing information into topics, todos, and decisions, it allows AI agents to 'grow' alongside the user. Whether you are managing complex software architectures or tracking daily progress, Memory Manager ensures your AI assistant remains informed and context-aware over weeks and months of development.

Memory Manager Use Cases

  • Automatically log every AI session to prevent the loss of critical technical insights.
  • Generate automated daily reports at 20:00 to review progress and pending tasks.
  • Extract 'evergreen' knowledge into a long-term MEMORY.md file via weekly distillations.
  • Capture real-time milestones such as skill deployments or financial transactions immediately.
  • Maintain a clean workspace by automatically purging expired temporary session files.

How Memory Manager Works

  1. Interaction Capture: The system monitors interactions and saves structured data (topics, decisions, emotion) into temporary session files.
  2. Daily Consolidation: Every evening at 20:00, the manager processes all temporary files to generate a unified daily summary.
  3. Weekly Distillation: On Sundays, the system analyzes the last seven days of summaries to update the primary long-term memory file.
  4. Real-time Injection: Critical events bypass the batch process and are appended directly to the permanent memory for immediate persistence.
  5. Automated Maintenance: The system executes a cleanup routine to delete temporary logs older than 30 days, keeping the storage footprint optimized.

Memory Manager Setup

To integrate this memory system into your workflow, ensure you have Python installed and follow these steps:

# Install the skill via the hub
clawhub install memory-manager

# Initialize the configuration
cp config.example.json config.json

You can then import the recording functions directly into your scripts or allow the agent to manage the lifecycle via scheduled tasks.

Memory Manager Data Schema & Taxonomy

Memory Manager organizes data across three tiers to ensure high retrieval accuracy:

File Type Location Description
Session Logs temp/session_*.md Raw interaction data including todos and decisions.
Daily Summaries memory/YYYY-MM-DD.md Batch processed summaries of a single day's activity.
Long-term Memory MEMORY.md The core repository for distilled, permanent knowledge.
Configuration config.json Defines directory paths and retention periods (default 30 days).

Memory Manager Advanced Features

  • Multi-agent support for shared team memory and collaborative knowledge bases.
  • Real-time 'Critical Node' triggers for immediate persistence of high-value data.
  • Configurable automated cleanup cycles to manage local storage overhead.
  • Structured metadata extraction including decision tracking and emotional context analysis.
  • Private deployment options for enterprise-grade data sovereignty within your Openclaw Skills ecosystem.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Bins python
Github Stars: 0
forks: 0

Featured*