A specialized automation skill that compresses daily interaction logs into concise long-term memories and archives raw files every morning.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install memory-compact
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install memory-compact 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 Compact?
Memory Compact is a high-performance utility designed for the Openclaw Skills ecosystem to solve the problem of context bloat and data fragmentation. By running a daily scheduled routine, it parses recent dialogue records, identifies critical decisions or preferences using keyword matching, and synthesizes them into a centralized long-term memory file.
This skill ensures your AI agent remains lean and focused by moving verbose historical logs to a dedicated backup directory while maintaining a high-level summary of important events. It is built with a security-first approach, operating entirely within local file boundaries without external network dependencies.
Memory Compact Use Cases
- Automatically summarizing daily AI interactions into a single MEMORY.md file.
- Archiving historical conversation logs to prevent workspace clutter.
- Generating daily status notifications for platforms like Feishu/Lark.
- Maintaining a clean, searchable timeline of key decisions made during development cycles.
How Memory Compact Works
- The system triggers the script at 06:30 Asia/Shanghai time via a configured cron job.
- The skill reads the most recent daily memory file (YYYY-MM-DD.md) from the workspace.
- It executes a keyword-based extraction process to identify 2-3 critical points based on terms like 'decision', 'important', or 'plan'.
- The extracted points are appended to the MEMORY.md file in a streamlined format.
- The original raw log is moved to the backup/memory directory for long-term storage.
- A summary notification is generated for external reporting.
Memory Compact Setup
To manually trigger the memory compression or view logs, use the following commands within your environment:
# Run the manual compression script
python3 /root/.openclaw/workspace/skills/memory-compact/wrapper.py
# Monitor the backup execution logs
tail -f /root/.openclaw/workspace/scripts/memory_backup.log
# Verify the extracted long-term memories
cat ~/.openclaw/workspace/MEMORY.md
Memory Compact Data Schema & Taxonomy
The skill manages data within the Openclaw Skills workspace using the following structure:
| Path | Description | Format |
|---|---|---|
memory/YYYY-MM-DD.md |
Source file for the current day's logs | Markdown |
backup/memory/ |
Directory for archived historical logs | Markdown Files |
MEMORY.md |
The master file containing synthesized long-term memory | Markdown |
memory_backup.log |
Technical execution logs for troubleshooting | Plain Text |
Memory Compact Advanced Features
- Customizable Extraction: Modify the
extract_key_points()function inmemory_backup.pyto add custom keywords or integrate LLM-based summarization. - Automated Scheduling: Built-in support for cron job integration with customizable time zones and execution payloads.
- Notification Integration: Automatically formats and prepares summary data for integration with Feishu (Lark) webhooks.
- Local Security Validation: Strict path verification mechanism that prevents operations outside of the authorized workspace.
SKILL.md
Loading
Related Openclaw Skills
Chinese Financial Regulation Monitor
An automated monitoring agent designed to track and summarize the latest policy updates and administrative penalties from China's primary financial regulatory bodies.
Free Models for AI Agents
A comprehensive utility for programmatically discovering, filtering, and selecting the most cost-effective models from OpenRouter for AI agent workflows.
ShortApi Video Models Aggregation Skill
A unified entry point to discover and integrate multiple high-end AI video generation models through a single API.
ShortApi Image Models Aggregation Skill
A unified gateway to discover, configure, and execute image generation tasks across multiple AI models using a single API endpoint.
Gold Price Report
A comprehensive utility for fetching real-time market data for gold, silver, and precious metals across spot, retail, and futures markets.
Creator Outreach
A specialized skill designed to craft high-conversion, personalized outreach sequences for creator partnerships and influencer marketing.
Featured*
