Home
Your AI Finally Remembers You
The only free, local-first AI memory system with universal architecture
🌐 Official Website • ⭐ Star on GitHub • ☕ Buy Me a Coffee • 💸 PayPal
SuperLocalMemory now learns your patterns and personalizes recall — all 100% locally. Adaptive ML ranking, workflow detection, and GDPR-compliant behavioral learning.
12 MCP tools | 7 skills | 6 resources | 2 prompts
- Learning System — How the three-layer learning architecture works
- Upgrading to v2.7 — Step-by-step upgrade guide
SuperLocalMemory is now production-hardened with trust enforcement, rate limiting, and accelerated graph building.
Feature Benefit Trust Enforcement Agents with trust < 0.3 blocked from write/delete operations Profile Isolation Full sandboxing — no cross-profile data leakage Rate Limiting Protects against memory flooding from misbehaving agents SSRF Protection Webhook dispatcher validates URLs to prevent SSRF attacks HNSW-Accelerated Graphs Faster knowledge graph construction at scale Hybrid Search Engine Combined semantic + FTS5 + graph retrieval v2.5 included: Real-time event stream, WAL-mode concurrent writes, agent tracking, memory provenance, 28 API endpoints.
Upgrade:
npm install -g superlocalmemory@latest
SuperLocalMemory V2 is an intelligent ai memory system that makes AI assistants like Claude, Cursor, and other mcp-server compatible tools remember everything about you and your projects. Created by Varun Pratap Bhardwaj, this local-first memory solution is the perfect alternative to Mem0 and Zep. Unlike cloud-based alternatives like Mem0 ($50+/mo) or Zep ($50/mo), SuperLocalMemory is:
- 100% Local — Your data never leaves your machine
- 100% Free — No usage limits, no credit systems, forever
- 100% Private — GDPR/HIPAA compliant by default
- Works Everywhere — 17+ IDEs including Claude Desktop, Cursor IDE, Windsurf, VS Code, and more
- Dual Protocol — MCP (agent→tool) + A2A (agent↔agent) support
| Guide | Description |
|---|---|
| Installation | 5-minute setup for Mac, Linux, Windows |
| Quick-Start-Tutorial | Your first memory in 2 minutes |
| CLI-Cheatsheet | Copy-paste command reference |
| Guide | Description |
|---|---|
| Universal-Architecture | 10-layer universal architecture with MCP + A2A integration |
| MCP-Integration | Model Context Protocol support for 17+ IDEs |
| Universal-Skills | 7 agent skills and slash-commands system |
| Knowledge-Graph-Guide | Auto-discovery of relationships |
| Pattern-Learning-Explained | How it learns your coding style |
| Multi-Profile-Workflows | Separate contexts for work/personal/clients |
| Guide | Description |
|---|---|
| CLI-Cheatsheet | Quick command reference |
| Python-API | Programmatic access |
| Configuration | Customization options |
| CHANGELOG | Version history and release notes |
| Guide | Description |
|---|---|
| Comparison-Deep-Dive | Detailed comparison with Mem0, Zep, Personal.AI |
| Why-Local-Matters | Privacy, GDPR, and local-first benefits |
| Guide | Description |
|---|---|
| FAQ | Frequently asked questions |
| Roadmap | Version history and planned features |
| Troubleshooting | Common issues and solutions |
| Contributing | How to contribute |
| Issues | Report bugs or request features |
Every Claude session:
You: "Remember that auth bug we fixed?"
Claude: "I don't have access to previous conversations..."
You: *sighs and re-explains everything*
# Save once
superlocalmemoryv2:remember "Fixed auth bug - JWT tokens expiring too fast"
# Recall forever
superlocalmemoryv2:recall "auth bug"
# ✓ Found: "Fixed auth bug - JWT tokens expiring too fast"| Feature | Mem0 | Zep | Personal.AI | SuperLocalMemory V2 |
|---|---|---|---|---|
| Price | Usage-based | $50/mo | $33/mo | $0 forever |
| Local-First | ❌ Cloud | ❌ Cloud | ❌ Cloud | ✅ 100% |
| IDE Support | Limited | 1-2 | None | ✅ 17+ IDEs |
| Universal Architecture | ❌ | ❌ | ❌ | ✅ MCP + Skills + CLI |
| MCP Integration | ❌ | ❌ | ❌ | ✅ Native |
| A2A Protocol | ❌ | ❌ | ❌ | 🔜 v2.8 |
| Pattern Learning | ❌ | ❌ | Partial | ✅ Full |
| Knowledge Graphs | ✅ | ✅ | ❌ | ✅ Leiden Clustering |
| Zero Setup | ❌ | ❌ | ❌ | ✅ 5-min install |
SuperLocalMemory V2 is the ONLY solution with universal IDE support, full local operation, and zero cost. Created by Varun Pratap Bhardwaj as an open-source alternative to expensive cloud services.
# Mac/Linux
git clone https://github.com/varun369/SuperLocalMemoryV2.git
cd SuperLocalMemoryV2 && ./install.sh
# Windows
git clone https://github.com/varun369/SuperLocalMemoryV2.git
cd SuperLocalMemoryV2; .\install.ps1That's it. No Docker. No API keys. No cloud accounts.
SuperLocalMemory implements cutting-edge 2026 research:
| Research | Source | Implementation |
|---|---|---|
| PageIndex | VectifyAI (Zhang et al., Sep 2025) | Hierarchical memory indexing |
| GraphRAG | Microsoft (Edge et al., 2024, arXiv:2404.16130) | Knowledge graph clustering |
| MemoryBank | Zhong et al., AAAI 2024 (arXiv:2305.10250) | Long-term memory for LLM agents |
| MACLA | Forouzandeh et al., Dec 2025 (arXiv:2512.18950) | Multi-agent collaborative learning |
| Hindsight | Latimer et al., Dec 2025 (arXiv:2512.12818) | Retrospective identity pattern learning |
| A-RAG | arXiv:2602.03442 | Multi-level retrieval |
Note: PageIndex attribution corrected from "Meta AI" to VectifyAI (Mingtian Zhang et al., Sep 2025). "xMemory (Stanford)" was a non-existent paper reference and has been replaced with three published research papers: MemoryBank, MACLA, and Hindsight.
The only open-source implementation combining all these research approaches.
Created by Varun Pratap Bhardwaj — Solution Architect & Original Creator
Building open-source tools that make AI assistants actually useful for developers. SuperLocalMemory V2.3.0 brings universal integration to 17+ IDEs while maintaining 100% local-first privacy.
If SuperLocalMemory saves you time:
- ⭐ Star on GitHub — helps others discover it
- ☕ Buy Me a Coffee — fuel development
- 💸 PayPal — direct support
- 💖 GitHub Sponsors — recurring support
100% local. 100% private. 100% yours.
Created by Varun Pratap Bhardwaj
🌐 Official Website: superlocalmemory.com 📦 NPM Package: npmjs.com/package/superlocalmemory 📖 Documentation: GitHub Wiki 💬 Support: GitHub Issues