EverOS
EverMind's local-first memory runtime for agents: conversations, files and trajectories stored as Markdown, indexed in SQLite and LanceDB, shared across coding tools.
Tags
Description
A Python library (pip install everos, PyPI 1.5.1, Python 3.12+; v1.5.1 released 2026-10-22). Markdown files are the source of truth: you can read, edit, diff and version them in Git, and a watcher re-syncs the local SQLite and LanceDB indexes. User episodes and profiles are kept apart from agent cases and skills, retrieval is scoped by user, agent, app, project and session IDs, and an editable Knowledge Wiki sits alongside memory. Integrations include an MCP server (uvx everos-mcp, EverMind-AI/EverOS-MCP) for Claude Code, Codex, Claude Desktop and Cursor, a Claude Code plugin and Dify tools; the MCP server works with the self-hosted runtime or EverOS Cloud. Apache-2.0.
Related
- Mem0: Memory layer for AI agents and apps: drop-in persistent memory that lets agents learn from past interactions.
- Basic Memory: Local-first MCP server that gives AI assistants persistent memory stored as Markdown files in a knowledge base you own.
- Hindsight: Open-source agent memory system from Vectorize focused on agents that learn: retain, recall and reflect operations over memory banks, with an MCP server and a managed cloud.
- LanceDB: Developer-friendly, open-source embedded retrieval library and multimodal lakehouse for AI.
Sources
- https://github.com/EverMind-AI/EverOS
- https://pypi.org/project/everos/
- https://github.com/EverMind-AI/EverOS-MCP
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)