Can I search through old memories semantically?

Yes, the skill supports semantic search via embeddings, allowing agents to retrieve relevant facts from long-term memory based on the meaning of a query.

Does it support different memory types?

Yes, it distinguishes between short-term (working memory), long-term (persistent facts), and knowledge-based (static or learned) memory types.

How does memory pruning work in this skill?

The skill uses time-based and size-based strategies to identify and remove low-importance or outdated memories, keeping the agent's context window clean and efficient.

What triggers the Memory Manager skill?

The skill is activated by phrases such as 'manage memory', 'optimize context', or 'handle agent memory' within the Claude Code environment.

What is elizaOS memory management?

It involves organizing an AI agent's short-term working memory and long-term persistent facts to ensure the agent remains contextually aware without exceeding token limits.

Memory Manager for elizaOS

Name: Memory Manager for elizaOS
Author: Dexploarer

byDexploarer

•

数据科学与机器学习

Optimizes AI agent memory usage and manages conversation context through advanced pruning and archiving strategies.

The Memory Manager skill is a specialized utility designed to enhance the performance and longevity of elizaOS agents by efficiently managing their memory architecture. It provides automated strategies for context window optimization, short-term and long-term memory consolidation, and semantic search capabilities. By implementing both time-based and size-based pruning, this skill prevents context overflow and ensures that agents retain the most relevant information, facilitating persistent and high-quality interactions across multiple sessions.

主要功能

01Short-term and long-term memory type management

022 GitHub stars

03Semantic memory retrieval using embedding-based search

04Automated memory pruning and context window optimization

05Importance scoring and conversation history archiving

06Memory decay implementation for temporal relevance

使用场景

01Optimizing context windows to reduce token usage and improve response latency

02Managing large-scale agent deployments within the elizaOS framework

03Maintaining long-term consistency in persistent AI agent conversations

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add dexploarer/hyper-forge memory-manager

For use in Claude.ai and ChatGPT

Download Skill