Open Source · MIT License

A living second brain that connects everything you think with

Your AI agents, your notes, and whatever comes next — unified in one knowledge graph. Open-source, local-first.

Your memory is scattered

Claude remembers you like concise code, but Cursor has no idea. You spent a week organizing design thinking in your notes, but the next time you discuss it with an AI, it knows nothing. Every tool quietly accumulates its own understanding of you, yet nothing connects them.

Siloed Memory

Each AI tool keeps its own memory. None of them talk to each other.

Flat Fact Lists

Existing AI memory just lists facts. No linking, no discovering, no evolving.

Growing Noise

As memories pile up, data grows redundant and harder to retrieve.

The deeper problem isn't data sync. It's cognitive fragmentation.

Everything a second brain should be

Not just storage — a genuine extension of your cognitive system.

Unified Memory

All your AI tools and notes share one knowledge graph. Conversations across different AIs, notes across different apps — no longer scattered in silos, but woven together.

Complete Profile

At the start of each conversation, your AI receives a profile distilled from all past interactions — who you are, how you think, what you care about.

Graph-Navigating Recall

Follow links between memories — from a decision to the discussion behind it, then to later changes it triggered. On-demand exploration, not keyword guessing.

Living Metabolism

Memories decay naturally, divergent scanning discovers hidden connections, and important patterns crystallize — like biological memory.

Note Sync

Import from Obsidian, Logseq, and Apple Notes with a single click. Notes and AI conversations cross-pollinate in one graph.

100% Local

All data lives in a single SQLite file on your machine, exportable to Markdown anytime. No cloud, no vendor lock-in. Fully local with Ollama or mix with cloud models.

Three cognitive operations

This isn't CRUD. It's modeled after how biological memory works — through the MCP protocol.

brain_prepare

Load memory context

Start of every conversation

Like opening a memory map — your AI receives your complete profile and key memories.

brain_recall

Navigate and retrieve

When you need historical context

More than search — AI follows links between memories, exploring the graph on demand. Each recall strengthens the memory.

brain_digest

Store new information

When something worth remembering emerges

Not just saving — incoming information is decomposed, connected, and woven into the existing knowledge network.

Background Metabolism

Like the brain organizing memories during sleep — TideMind continuously maintains the graph.

Natural Decay

Inactive memories fade. Hub nodes are protected.

Divergent Scanning

Discovers hidden connections between unlinked ideas.

◆

Crystal Emergence

Important patterns crystallize into distilled insights.

How it's different

	Typical AI Memory	TideMind
Profile	No profile, or siloed within a single AI	Complete profile built from all your data, delivered at the start of every conversation
Storage	Flat list of facts	Knowledge graph where memories dynamically evolve across multiple dimensions
Recall	Keyword / vector top-K	Starts from relevant memories, follows links to explore, expands on demand
Emerge	Stored memories stay frozen	A living system that surfaces new cognition
Data	Cloud / vendor-locked	Local SQLite file, export to Markdown anytime

See what your brain is thinking

The desktop client is an inspection tool — for observing what your external brain is discovering and forgetting. Your main interface is whatever AI tool you already use.

Dashboard

Real-time pulse of your second brain — metrics, activity feed, crystal discoveries.

Brain Explorer

Visualize how your memories connect across domains in an interactive graph.

Memory Detail

Four-dimensional maturity model, evidence chains, and linked nodes.

Works with your tools

Connect through the MCP protocol — one config, no habit changes.

AI Tools

Claude Code Native MCP support

Cursor MCP configuration

Windsurf MCP configuration

Codex MCP configuration

Any MCP Tool Standard protocol

Note Systems

Logseq File watching, incremental sync

Obsidian Vault import, Canvas support

Apple Notes Read-only sync

More coming...

Frequently Asked Questions

Common questions about TideMind — what it is, how it works, and how it's different.

How is TideMind different from Obsidian / Notion / Mem?

TideMind is not a note-taking app — it's a memory layer that sits between your tools. Obsidian and Notion are where you write; TideMind is where all your thinking connects. It ingests notes from Obsidian, Logseq, Apple Notes, and conversations from your AI tools, then weaves them into a single living knowledge graph. You keep using whatever tools you prefer.

I already have a note system. Why do I need TideMind?

Your notes are one slice of your thinking. The other slices — AI conversations, decisions made in passing, preferences expressed across tools — are scattered and siloed. TideMind captures all of them and builds connections you wouldn't make manually. It also delivers this context to your AI automatically, so every conversation starts with full awareness of who you are.

Where is my data stored? Is it secure?

Everything lives in a single SQLite file on your machine. No cloud, no third-party servers, no accounts. You can export all your data to Markdown at any time. If you want full privacy, run TideMind entirely offline with local models via Ollama.

Do I need an API key? Which models are supported?

TideMind works out of the box with zero API keys — basic text search and storage are fully functional. For advanced features like automatic extraction, memory reconsolidation, and crystal emergence, you can connect Anthropic, Google Vertex AI, or Gemini. Mix and match freely — e.g., Ollama locally for embeddings + Anthropic for LLM.

How do I install it? Do I need a technical background?

Two options: Desktop App (download, install, done — no terminal needed) or From Source (for developers: git clone, npm install, npm start). The desktop app handles all configuration through a visual settings panel.

"Memories naturally fade" — does that mean my data gets deleted?

No. The raw data is always preserved. What changes is relevance: memories you haven't revisited gradually lose prominence in retrieval, just like in a real brain. Important, highly-connected memories are protected from decay. Any faded memory is instantly reinforced the moment you or your AI recall it. It's prioritization, not deletion.

What's the difference between TideMind (open source) and TideMind Cloud?

The open-source version is fully functional for local use — all core features, no limitations. TideMind Cloud (coming soon) adds cross-device sync, mobile access, and a managed account system. Some features requiring cloud infrastructure (such as browser extension integration) may only be available in the Cloud version. The core engine is identical.

Get TideMind

Start building your external brain today. All your data stays on your machine.

Beta

macOS

Apple Silicon & Intel

Download DMG

Coming Soon

Cloud Sync

Cross-device sync & mobile

Stay Tuned

From Source

Node.js >= 18

Quick Start Guide