novyxlabs

novyxlabs / novyx-mcp

Public

Persistent memory for AI agents. 64 MCP tools for remember, recall, rollback, audit, knowledge graph, governed actions, context spaces, and execution tracing. pip install novyx-mcp

novyxlabs.com agents ai-agents audit-trail claude claude-code
16
3
89% credibility
Found Mar 22, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Novyx MCP adds persistent shared memory capabilities to AI agents in tools like Claude Desktop, Cursor, and Claude Code, supporting local storage or cloud syncing with features for recall, knowledge graphs, rollbacks, and multi-agent collaboration.

How It Works

1
📖 Discover shared memory for AIs

You learn about a helpful tool that lets your AI assistants like Claude remember and share important details across different chats and projects.

2
🛠️ Add the tool easily

You install it on your computer in just a few seconds so it's ready to use.

3
Pick your setup style
🏠
Local mode

No sign-up needed—your memories stay safe on your machine with full features.

☁️
Cloud mode

Sign up free online for extra powers like sharing with others and advanced debugging.

4
⚙️ Link it to your AI app

You add a quick note in your AI tool's preferences, like Claude or Cursor, to turn on the memory sharing.

5
📝 Guide your AIs

You add a simple instruction to your project's notes file, telling AIs to check and save shared memories.

6
✨ Watch the magic

In your next AI chat, it pulls up memories from past sessions, feeling like your helpers are truly working together.

🎉 AIs become a team

Now every AI session builds on the last, sharing knowledge, decisions, and progress for smarter, coordinated results.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 16 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 novyx-mcp?

novyx-mcp delivers persistent memory for AI agents, letting you store observations, build knowledge graphs, and share context across sessions in Claude Desktop, Cursor, or Claude Code. It solves the LLM amnesia problem—agents forget everything between runs—via 64 MCP tools for remembering, recalling, rolling back changes, auditing actions, and creating context spaces for multi-agent teams. Pip install it, add a config snippet, drop a CLAUDE.md note, and your agents gain github persistent storage with semantic search, replay debugging, and execution tracing, all in Python over the Model Context Protocol.

Why is it gaining traction?

Unlike basic key-value caches, it offers rollback previews, cryptographic audits, and time-travel replays for debugging agent drifts, plus local SQLite mode for zero-config persistent memory llm workflows versus cloud for team sharing. Developers hook on the CLAUDE.md integration that auto-coordinates isolated sessions into persistent empires of agents handling governed actions. Semantic recall with tags and scores beats naive keyword search, and prompts like memory-context inject relevant history on demand.

Who should use this?

AI engineers building multi-agent systems in Claude Code who need shared persistent memory across sessions, like coordinating research bots or dev workflows. Teams experimenting with persistent memory programming for LLMs in Cursor, tired of re-explaining context every run. Claude Desktop users wanting knowledge graphs and audit trails for agent actions without custom backends.

Verdict

Grab it for Claude agent projects needing persistent memory module—local mode works out of the box, cloud adds polish—but with 16 stars and 0.9% credibility score, it's beta-raw; test locally before production. Solid docs and MIT license make it low-risk to prototype.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.