OpenLore

OpenLore turns any evolving codebase into a navigable knowledge graph backed by OpenSpec living specifications. It maintains persistent architectural context across AI agent sessions, reducing token consumption for orientation from 15,000-50,000 to about 1,000-3,000 tokens. The three-layer system includes static analysis, spec generation, and an MCP agent runtime with 45 tools.

OpenLore sits in the RAG / Knowledge Base category, so it is usually better judged by workflow fit, deployment style, and team needs than by isolated features alone. Based on the available metadata, it leans most clearly toward Persistent call graph with incremental updates via --watch-auto for live codebase awareness. A representative use case is AI coding agents need to quickly understand codebase structure and maintain context across sessions without re-reading files. The listed license is MIT. GitHub stars are currently 350. Relevant tags include adr, agentic-workflows.