GuyMannDude

GuyMannDude / mnemo-cortex

Public

Open-source memory coprocessor for AI agents. Persistent recall, semantic search, crash-safe capture. No hooks required.

15
7
100% credibility
Found Mar 11, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Mnemo Cortex is a drop-in memory system for AI agents that captures conversations, provides context retrieval, and validates responses to enable persistent recall across sessions.

How It Works

1
🔍 Discover Mnemo Cortex

You hear about a helpful tool that fixes your AI assistant's forgetfulness between chats.

2
📦 Get it ready

You add the memory helper to your computer in just a few moments.

3
🧙 Friendly setup guide

A simple wizard walks you through connecting your AI's thinking power, picking what feels right.

4
▶️ Turn it on

You start the quiet background service that watches and remembers everything.

5
👀 Watch conversations

You activate auto-save so every chat gets captured without you lifting a finger.

6
🧠 Magic recall happens

Now your AI pulls up perfect memories from past talks instantly.

🚀 AI never forgets

Your assistant remembers details forever, making work smooth and smart.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 13 to 15 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 mnemo-cortex?

Mnemo Cortex is a Python-based open source memory coprocessor for AI agents, solving the amnesia problem where LLMs forget everything between sessions. It runs as a lightweight server on port 50001 with four HTTP endpoints: retrieve relevant context via semantic search, validate draft responses with PASS/ENRICH/WARN/BLOCK, ingest live prompt-response pairs for crash-safe capture, and archive sessions. Install via pip or Docker, configure providers like Ollama or OpenAI through a CLI wizard, and use watchers to auto-capture conversations from frameworks like OpenClaw—no code hooks required.

Why is it gaining traction?

This stands out as an open source LLM with memory and open source memory for AI agents, offering multi-tenant isolation, persona modes (strict for facts, creative for brainstorming), and a hot/warm/cold storage lifecycle that keeps recent exchanges instantly searchable. Developers hook it up via simple HTTP calls or adapters, with resilient fallbacks if a local model like Ollama fails over to cloud APIs. The CLI handles everything from init to status checks, making it a drop-in layer for persistent recall in agent workflows.

Who should use this?

AI agent builders using OpenClaw, Agent Zero, or custom LLM setups who lose context across runs. Solo devs prototyping multi-session agents for tasks like project tracking or customer support bots. Teams wanting a self-hosted open source memory layer without vector DB complexity or cloud vendor lock-in.

Verdict

Try it if you're building agents—solid docs, 56 tests, and MIT license make it easy to extend, but with only 13 stars and 1.0% credibility score, it's early beta so expect rough edges in production. Pairs well with local Ollama for zero-cost memory testing.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.