Skip to content
repository radar
RAG & Memory

🧠memU

ACTIVE BREAKOUT

NevaMind-AI/memU · homepage ↗

Personal memory across agents

⭐ Very popular: 14k stars, gaining about 52 a week

View on GitHub ↗

repo profile

vintage 2025 1 year old
delivery product
language Python
license Other

momentum

total stars 14k
stars added last week +52
weekly downloads 274 /wk · PyPI
binary downloads 228 total · GitHub releases
commits / week 8 cooling · -60% vs prior mo
issues closed 71% ~18d to close, median
contributors 41 (+1 in 15d)
release cadence monthly
last activity 3d ago

durability

backing community / independent independent
bus factor 2 concentrated
top-author share 28% 6 mo

bus factor = how many people it takes to cover more than half the commits (6 months). 1 is a solo project; higher means the work is spread across a team. top-author share is the single busiest author's slice of those commits.

since we covered it

monthly average + 890/mo (+10%/mo) · + 6k total since PR#27

why it's a big deal

  • Gives AI agents persistent, per-user memory so they can pull scoped, relevant context instead of rescanning full histories or stuffing long data into every prompt.
  • Helps developers building agents cut token cost and latency by retrieving compact context, and it stores everything as human-readable Markdown that people can inspect and edit.
  • Automatically extracts reusable tool patterns and workflows from an agent's tool traces, so an agent improves its skills over time without manual retraining.

under the hood

  • Written in Python and organized as a three-layer Markdown structure, an INDEX file mapping memories, a MEMORY file for facts and preferences, and a SKILL file for extracted tool patterns.
  • Offers two retrieval modes, vector-first recall with optional LLM routing, and LLM-ranked recall at each tier, exposed through memorize and retrieve APIs.
  • Supports in-memory, SQLite, or PostgreSQL with pgvector as storage backends, and connects to OpenAI, OpenRouter, DashScope, and Voyage AI for models.

Radar summary, generated from the project's public sources

star history

PR#27 · 9k14k now Jul 2025Aug 2026
  1. PR#27 9k 2026-02-11
  2. now 14k + 6k since first covered

curve is sampled from GitHub's star history, plus our own daily readings since we covered it; the dashed stretch is before we first covered it, the solid line since. figures at coverage are the numbers we printed then (approx.), current count is live.

understory

Better known than its recent output, coasting a little on attention.

-13 understory score output 45 · clout 58
Aug 2025 Jul 2026
  • output, commits & releases
  • clout, star velocity

output = commits & releases; clout = star velocity, both 0 to 100 monthly indices; the gap where output runs above clout is the understory. The understory →

covered in

  • PR#27 2026-02-11 below the radar

    Memory for 24/7 proactive agents

similar projects

compare these →
  • 🧠 Memori

    Memori is agent-native memory infrastructure. A LLM-agnostic layer that turns agent execution and conversation into structured, persistent state for production systems. Built for enterprise, Memori works with the data infrastructure you already run, no rip-and-replace, and deploys across managed cloud, single-tenant cloud, VPC, and on-premises.

    16k ACTIVE
  • 🧩 cc switch

    Rust · 8.9× the stars

    A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io

    127k ACTIVE
  • 🗺️ Understand-Anything

    TypeScript · 5.5× the stars

    Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.

    79k ACTIVE

comments

Sign in with GitHub to add your blip on memU.

loading comments…