SOAT

关于

SOAT (Source of Agentic Truth) is an open-source framework designed to equip autonomous AI agents with essential persistent memory capabilities. It enables agents to store, retrieve, and manage information effectively across sessions, leveraging semantic search powered by vector embeddings. This framework addresses the critical need for AI agents to maintain context and recall past interactions reliably, offering native Model Context Protocol (MCP) integration and a simple REST API for broad compatibility.

主要功能

• Persistent Memory for AI agents
• Semantic Recall using natural language queries
• Vector Search powered by pgvector
• Model Context Protocol (MCP) server integration
• Simple REST API for universal application integration
• 1 GitHub stars

使用案例

• Enabling AI agents to maintain conversational context and recall past interactions
• Enhancing AI agents with long-term memory across sessions
• Providing a memory layer for autonomous AI applications requiring semantic data retrieval