This project provides a framework for implementing cognitive architecture and psychological memory concepts into agentic LLM systems, addressing the stateless nature of LLMs. It targets developers building sophisticated LLM agents that require more than simple prompt-based context, enabling agents to exhibit human-like memory and learning capabilities.

How It Works

The project models four distinct memory systems within a Retrieval-Augmented Generation (RAG) agent: Working Memory for immediate context, Episodic Memory for past experiences, Semantic Memory for factual knowledge, and Procedural Memory for interaction rules and skills. This layered approach aims to imbue LLM agents with a more holistic cognitive design, allowing them to learn from and apply past learnings to new situations.

Quick Start & Requirements

• Install: pip install -r requirements.txt
• Prerequisites: Python 3.x, OpenAI API key.
• Demo: https://github.com/ALucek/agentic-memory/blob/main/Agentic%20Memory.ipynb

Highlighted Details

• Implements four memory types: Working, Episodic, Semantic, and Procedural.
• Utilizes a RAG agent architecture.
• Focuses on cognitive architecture for LLM agents.

Maintenance & Community

• Maintained by ALucek.
• No community links (Discord, Slack) or roadmap are provided in the README.

Licensing & Compatibility

• The README does not specify a license.
• Compatibility for commercial use or closed-source linking is undetermined.

Limitations & Caveats

The project is presented as a conceptual notebook implementation, and its readiness for production deployment or robustness in complex scenarios is not detailed. The lack of explicit licensing also poses a barrier to commercial adoption.