memU Model Card
这是可索引项目报告证据页：它保留 memU Model Card 的源材料入口、机制线索和限制提醒；正文仍需 reader/editor 与 academic public-copy review 后才能当作最终结论引用。
memU Model Card
| Field | Value |
|---|---|
| Repository | NevaMind-AI/memU |
| Category | 24/7 Proactive Agent Memory |
| Stars / forks snapshot | 13700 / 1000 |
| Language | Python |
| License | Apache-2.0 |
| Raw capture | raw-github/nevamind-ai_memu.md |
| Updated by | hourly public metadata update, 2026-05-24 |
1. Role in Self Evolve
memU 是面向 24/7 proactive agents 的 memory framework，公开页面把它定位为 OpenClaw-like always-on assistant 的长期记忆底座，关注低 token 成本、用户意图理解和主动行动。
2. Working Principle
always-on interaction stream -> hierarchical memory filesystem -> intent inference -> proactive agent action
3. Evidence Path
web GitHub page observed 288 commits, Apache 2.0 license, Python primary language, OpenClaw alternative section, custom LLM/provider docs, 13.7k stars and 1k forks. Shell GitHub API access remained blocked by DNS and local gh auth was invalid, so this card treats the current snapshot as web-observed rather than API-verified.
4. Teaching Use
Use this card to explain Proactive agent memory substrate in the raw -> classification -> project card -> site/report pipeline. The reading path is: raw capture -> classification row -> public site card -> project report -> aggregate GitHub analysis.
5. Limits
当前未克隆源码，未运行 benchmark、SDK examples 或 skill install flows；star/fork/commit 快照来自公开 GitHub 页面文本或可见页面片段。