Skip to content

alibaizhanov/mengram

AI Agents & Assistants

Human-like memory layer for AI agents with semantic, episodic, and procedural memory. Claude Code hooks (auto-save, auto-recall, cognitive profile). 29 MCP tools, knowledge graph, smart triggers, multi-user isolation. Python & JS SDKs.

Python Latest v2.25.3 · 2mo ago Security brief →

Features

  • Provides semantic memory for facts and preferences
  • Stores episodic memory (events and decisions)
  • Maintains procedural memory that evolves from failures
  • Supports native multilingual retrieval across 23 languages

Recent releases

View all 6 releases →
Upgrade now
v2.25.3 Breaking risk
Breaking upgrade Dependencies

Agent-native install

Config change
v2.23.0 New feature
Auth Breaking upgrade

Email‑verification bypass + health endpoint

No immediate action
v2.22.0 New feature

Quota usage headers

No immediate action
v2.21.2 New feature

LangChain integration update

No immediate action
v2.21.0 Breaking risk

Agent mode

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

About

Stars
183
Forks
26
Languages
Python HTML JavaScript
Downloads/week
36 ↓52%
NPM Maintainers
1 Single npm maintainer
Contributors
5
TypeScript
Types included ✓

Install & Platforms

Install via
pip npm

Alternative to

claude-mem Mem0 Zep Letta

Beta — feedback welcome: [email protected]