# EverMind-AI/EverOS

One portable memory layer for every AI agent: local-first, Markdown-native, user-owned, and self-evolving across apps, tools, and workflows.

Repository: https://github.com/EverMind-AI/EverOS
Canonical: https://ross.abutalabs.com/products/everos
Homepage: https://evermind.ai/everos
Language: Python
License: Apache-2.0
License Family: permissive
Topics: agentic-ai, ai, chats, llm, long-term-memory, memory, memory-management, python3, rag, agent-memory, clawdbot, clawdbot-skill, mcp, skills, deepseek-harness, dsh, dsh-plugin
Last push: 2026-08-26T07:17:34+00:00

## Health v2 (maintenance only)
Score: 83/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 96, longevity 22
- inputs: {"age_days": 309, "days_push": 7, "days_rel": 26, "gap_med": 5.5, "n_releases_24m": 11}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12452, forks 903 (observed 2026-08-28T04:10:52.721518+00:00)

## What it is
EverOS is a Python library and local-first memory runtime that gives AI agents a portable, user-owned memory layer across apps, tools, and workflows. It stores conversations, files, and agent trajectories as readable Markdown files, backed by local SQLite and LanceDB indexes for fast retrieval and self-evolving skill reuse.

## Use cases
- give my AI agent long-term memory across sessions
- store chat history and agent trajectories as editable markdown files
- add persistent memory to a Claude Code or MCP-compatible agent
- let agents learn reusable skills from completed tasks
- run a local-first memory layer without MongoDB or Redis
- search agent memories by user, agent, and context
- sync memory across multiple AI apps and devices

## When to choose
- you want local-first, Git-versioned Markdown as the source of truth for agent memory
- you need cross-agent, cross-platform memory without managed database services
- you want self-evolving procedural memory (cases promoting into skills)
- you need MCP, OpenAI SDK, or Anthropic SDK compatibility

## When to avoid
- you need a fully managed cloud memory service with zero local setup
- you require graph-based memory or entity-relation modeling out of the box
- your stack is not Python-based and you cannot use the SDK or MCP interface

## Facets
- artifact type: library
- maturity: active
- function: rag, vector-database, database, mcp, agent-framework, search-engine, sdk
- domain: artificial-intelligence, large-language-models, developer-tools, databases
- platform: python, cross-platform, self-hosted, cli
- tags: agent-memory, long-term-memory, markdown-native, local-first, lancedb, sqlite, self-evolving-skills, memory-management, procedural-memory, mcp-server, ai-agents, retrieval-augmented-generation

## Member repositories
- EverMind-AI/EverOS (main) score 83

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:52.721518+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:14:03.301334+00:00, confidence not recorded.
  - readme: https://github.com/EverMind-AI/EverOS (fetched 2026-08-28T04:10:52.721518+00:00, sha 29b405336579)
  - homepage: https://evermind.ai/everos (fetched 2026-08-29T08:10:49.868710+00:00, sha 1293e0e382cc)
  - registry_pypi: https://pypi.org/pypi/everos/json (fetched 2026-08-29T08:10:49.878403+00:00, sha ce87b9ddd350)
- Data as of 2026-08-30T08:39:29.467469+00:00.