Kindex is a persistent knowledge graph tool designed to provide AI coding agents with enduring, context-aware memory. By integrating with various Model Context Protocol (MCP)-capable agents such as Claude Code, Codex, Gemini CLI, and others, Kindex indexes conversations, projects, and intellectual work. This enables agents to start each session with relevant context, understanding relationships and constraints, and managing context injection efficiently based on token budget. Kindex is available as a free MCP plugin or as a standalone CLI, making it suitable for developers, teams, and enterprises seeking to enhance AI workflow memory and context-awareness.
Visit Kindex's official website for product details and getting started.
Comprehensive guides and API references for integrating and using Kindex effectively.