# MemTensor/memmy-agent

🍙  A personal AI agent & local memory hub for all AI agents, gives every AI one shared, fully controlled memory and persistent context — all AI remember the same you. Now supports Claude Code, Codex, OpenClaw and Hermes Agent etc.

Repository: https://github.com/MemTensor/memmy-agent
Canonical: https://ross.abutalabs.com/products/memmy-agent
Homepage: https://memmy.bot
Language: TypeScript
License: MIT
License Family: permissive
Topics: agent, ai-agents, memory, agentic-ai, llm, long-term-memory, personal-ai, chatgpt, claude-code, cli, codex, cursor, desktop-app, desktop-pet, hermes-agent, mcp, openai-api, openclaw, tui, dsh-plugin
Last push: 2026-09-02T14:32:39+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 99, longevity 3
- inputs: {"age_days": 48, "days_push": 0, "days_rel": 7, "gap_med": 4.0, "n_releases_24m": 11}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1232, forks 109 (observed 2026-09-03T02:15:06.632259+00:00)

## What it is
Memmy is a personal AI agent and local memory hub that gives every AI agent (Claude Code, Codex, Cursor, OpenClaw, etc.) a shared, locally controlled long-term memory and persistent context. It ships as a desktop app, CLI, and HTTP/OpenAI-compatible API, powered by a MemOS-based memory engine with local SQLite storage.

## Use cases
- share one memory across Claude Code, Codex, and Cursor
- continue unfinished tasks when switching between AI coding agents
- turn scattered AI chat history into structured personal memory
- remember my coding preferences so I never repeat them to new agents
- hand tasks directly to a personal AI agent that knows my context
- inject relevant project context into whichever agent I'm using
- run a local memory service for external agents via CLI or API
- consolidate weekly technical decisions into a document automatically

## When to choose
- you work with multiple AI agents daily and want shared, persistent context
- you want local-first memory storage with BYOK model support
- you need cross-agent task continuity without re-explaining your project
- you want an agent runtime with MCP, scheduled tasks, and skills built in

## When to avoid
- you need a purely cloud-hosted memory service with team collaboration
- you only use a single AI tool and don't need cross-agent memory
- you require fully offline operation without any account or trial tokens
- you need a lightweight library to embed memory in your own app rather than a standalone app

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, mcp, rag, chatbot, cli, search-engine, sdk
- domain: large-language-models, developer-tools
- platform: windows, cli, cross-platform
- tags: long-term-memory, memory-hub, cross-agent-context, desktop-pet, local-first, byok, claude-code, codex, cursor, openclaw, memos, sqlite, tui, task-continuity, ai-agents, automation, macos, desktop, typescript

## Member repositories
- MemTensor/memmy-agent (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:06.632259+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 2026-08-30T07:12:23.606197+00:00, confidence not recorded.
  - readme: https://github.com/MemTensor/memmy-agent (fetched 2026-09-03T02:15:06.632259+00:00, sha 585fe757016c)
  - homepage: https://memmy.bot (fetched 2026-08-29T13:12:32.997794+00:00, sha dee83b903674)
  - site_page: https://memmy.bot/docs (fetched 2026-08-29T13:12:33.006875+00:00, sha 16ebde9f7fe4)
- Data as of 2026-08-30T08:39:29.467469+00:00.