关于
AI Agent History RAG solves the common compaction problem faced by AI coding agents, where long sessions often lead to a loss of context. By acting as a Model Context Protocol (MCP) server, it continuously ingests and indexes the history from various AI coding tools like Claude Code, Codex, Gemini CLI, and Google Antigravity. This creates a centralized, searchable vector database, enabling users to semantically query past conversations, code changes, and tool outputs. It supports both single-machine local operation and multi-machine client-server deployments, ensuring context is never lost and is accessible across all development environments.
