A sophisticated memory management system that scores, prunes, and archives AI agent records to maintain high-relevance context and optimal token efficiency.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install daxiang-memory-optimization
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 daxiang-memory-optimization 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 Optimization Skill?
The Memory Optimization Skill is designed to solve the problem of context bloat in AI agents by implementing a multi-level memory architecture ranging from raw logs (L0) to core insights (L3). By utilizing a weighted scoring algorithm that considers keyword matching, temporal decay, and access frequency, it ensures that your agent always operates with the most pertinent information. This skill is a core component for developers looking to build scalable and cost-effective agents within the Openclaw Skills ecosystem.
Memory Optimization Skill Use Cases
- Reducing LLM token costs by automatically archiving low-value conversation history.
- Improving agent response accuracy by prioritizing high-relevance context during retrieval.
- Transitioning raw daily logs into structured knowledge bases and SOPs.
- Managing long-term projects where maintaining a lean, high-density memory window is critical.
How Memory Optimization Skill Works
- Every memory entry is assigned a relevance score (0.0-1.0) based on keywords, time decay, access frequency, and tag matching.
- An automatic pruning mechanism identifies entries falling below a specific threshold (default 0.6) and moves them to an archive.
- The system enforces a capacity window (default 200 entries) to prevent context overflow.
- Smart retrieval logic re-ranks memories during queries to provide the agent with the most significant data points.
- Periodic maintenance routines perform incremental scoring and archive cleanup to ensure long-term system health.
Memory Optimization Skill Setup
To deploy this skill within your workspace, follow these steps:
- Initialize the required directory structure:
mkdir -p memory/archive memory/summaries library/insights library/sops
- Create the configuration file
memory-optimization-config.jsonin your root directory to define your scoring weights and pruning thresholds. - Integrate the Python or PowerShell scoring functions into your agent's memory lifecycle hooks.
Memory Optimization Skill Data Schema & Taxonomy
The skill utilizes a hierarchical metadata taxonomy to organize information across four distinct levels:
| Level | Storage Location | Content Type | Retention |
|---|---|---|---|
| L0 | memory/YYYY-MM-DD.md |
Raw logs and full session history | 100% Data |
| L1 | memory/summaries/ |
Daily key points and decisions | 70% Token Reduction |
| L2 | MEMORY.md |
Structured long-term knowledge | High Value |
| L3 | library/insights/ |
Core SOPs and refined references | Top 5% Essence |
Memory Optimization Skill Advanced Features
- Incremental Scoring: Minimizes overhead by only recalculating scores for new or modified memory entries.
- Cross-Platform Implementation: Provides full logic support for both Python-based and PowerShell-based environments.
- Relevance Caching: Implements a TTL-based cache for scores to accelerate high-frequency retrieval tasks.
- Custom Decay Curves: Allows developers to configure how quickly information loses relevance over time based on specific use cases in Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Job Autopilot - Resume Tailor
An AI-powered skill that automatically tailors resumes and cover letters to specific job descriptions while maintaining 100% factual accuracy.
Supervising Agents Skill
A professional framework for AI agents to dispatch, monitor, and verify subagent tasks through proactive intervention and evidence-based reporting.
SpielerPlus Scraper
A specialized scraper for the SpielerPlus and TeamPlus platforms designed to extract team events, member data, and financial records.
Slack Automation Skill
A specialized browser automation skill designed to interact with Slack workspaces for messaging, data extraction, and workflow automation.
Job Autopilot Search Agent
An automated recruitment scout that builds a candidate profile from your resumes to discover, filter, and track relevant job openings across the web.
Dogfood: Automated QA & Exploratory Testing
Dogfood provides a systematic way to explore web applications and find bugs, producing reports with full reproduction evidence as one of the essential Openclaw Skills.
Featured*
