alibaizhanov

alibaizhanov / mengram

Public

Human-like memory for AI agents — semantic, episodic & procedural. Experience-driven procedures that learn from failures. Free API, Python & JS SDKs, LangChain & CrewAI integrations.

mengram.io ai-agents ai-memory cognitive-profile crewai episodic-memory
66
8
100% credibility
Found Feb 19, 2026 at 24 stars 3x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Mengram is a memory system for AI agents that captures facts, events, and self-improving workflows from conversations, offered as free open-source software with cloud hosting and integrations.

How It Works

1
🌐 Discover Mengram

You hear about a smart memory tool that helps AI friends remember chats, facts, and helpful steps like a real person.

2
📝 Sign up easily

Visit the website, enter your email, and get a free personal space for your AI's memories in seconds.

3
🧠 Share your first story

Paste a conversation or note, and watch it instantly organize facts, past events, and useful routines.

4
🔍 Find what you need

Ask natural questions like 'what went wrong last time?' and get tailored answers from your saved memories.

5
🤖 Link to your AI buddy

Connect it to your AI helper with a simple copy-paste, so it remembers you across every chat.

✨ AI grows wiser

Your AI now recalls details, repeats what works, and improves routines automatically from real experiences.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 24 to 66 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is mengram?

Mengram builds human-like memory for your AI agents, blending semantic facts, episodic events, and procedural workflows that evolve from failures. It solves the stateless agent problem by storing experiences—like deployment crashes or customer interactions—and auto-improving procedures, unlike rigid fact databases. Python core with free cloud API, JS SDK, LangChain/CrewAI tools, and MCP server for Claude Desktop integration.

Why is it gaining traction?

Unique experience-driven evolution: feed in failures, and procedures version and adapt automatically, bridging episodic memory to actionable steps. Unified search pulls from all three memory types, plus cognitive profiles generate ready system prompts for personalized LLMs. Free API beats paid Mem0/Letta tiers, with quick CLI setup for local testing.

Who should use this?

CrewAI/LangChain builders needing agents that remember past runs and learn workflows from failures. DevOps teams scripting deployment agents with human-like episodic memory for infinite-context LLMs. Indie devs prototyping multi-session AI assistants via the REST API or JS SDK.

Verdict

Solid pick for integrating dynamic human-like memory recall in LLM agents—procedural learning stands out. With 16 stars and 1.0% credibility, it's pre-1.0 maturity; docs shine but expect rough edges. Prototype with the free API key now if agent persistence is your blocker.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.