TideMind

A living second brain that connects everything you think with — your AI agents, your notes, and whatever comes next.

English | 中文

Why TideMind

Your memory is scattered across a dozen tools, and none of them talk to each other.

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.

The deeper problem: existing AI memory is just listing and storing facts. It won't link memories together, won't discover commonalities across your ideas that you haven't noticed yourself, and certainly won't build a complete profile of who you are. And as memories pile up, the data grows increasingly redundant and harder to retrieve.

TideMind is a memory layer that spans all your tools — a living knowledge graph that connects your AI, your notes, and your thinking.

What is TideMind

All your memory, in one place. TideMind connects your AI tools, your notes, and whatever comes next into a single knowledge graph. Conversations across different AIs, notes across different apps — no longer scattered in silos, but woven together, cross-pollinating.

Every AI knows you completely. At the start of each conversation, TideMind provides your AI with a profile and key memories distilled from all past interactions — who you are, how you think, what you care about.

More precise, more complete recall. When your AI needs more context, it doesn't just rely on keyword/vector search — it can follow links between memories, navigating from one decision to the discussion behind it, then to later changes it triggered. No need to load everything; no risk of missing what matters. It starts with a few highly relevant nodes and explores along links when more context is needed — on-demand exploration.

More than a database — a living system. Memories form connections automatically. Unimportant details naturally fade; important ones are reinforced and updated. TideMind actively discovers commonalities you haven't noticed — linking a decision from three months ago to the problem you're facing today. Information doesn't end at storage. It's digested, connected, distilled, and sometimes forgotten. That's what "living" means.

Simple to start. Connect your AI tools through the MCP protocol — one config, no habit changes. Import and sync notes with a single click.

Your memory belongs to you. All data lives in a single SQLite file on your machine, exportable to Markdown anytime. No cloud, no vendor lock-in. Go fully local with Ollama for privacy, use cloud models for quality, or mix both.

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 to AI at the start of every conversation
Storage	Flat list of facts	Knowledge graph where memories dynamically evolve across multiple dimensions over time and use
Recall	Keyword / vector top-K	Starts from the most 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

Quick Start

Option 1: Download the Desktop App

The easiest way to get started. Download, install, and you're ready to go.

Download for macOS (Apple Silicon) — Beta

Open the app, then head to Settings to connect your AI tools and note systems with one click.

Option 2: From Source

For developers who prefer building from source.

Prerequisites

• Node.js >= 18

That's the only hard requirement. TideMind runs out of the box with text search (BM25) and basic storage.

Optional, for the full experience:

• An LLM provider (Anthropic API Key, Google Vertex AI, or Gemini) — enables automatic information extraction, memory reconsolidation, divergent scanning, and crystal emergence. Without it, these features are simply skipped.
• An embedding provider (Ollama, Vertex AI, or Gemini) — enables semantic vector search alongside BM25. Without it, search falls back to keyword matching.
• You can mix and match: e.g., Ollama locally for embeddings + Anthropic for LLM, or go fully local, or fully cloud. See the Integration Guide for details.

Install

git clone https://github.com/SawyerHan-AI/tidemind.git
cd tidemind
npm install
npm run build

Launch and Connect

npm start

Open the desktop client, then head to Settings to connect your AI tools and note systems with one click — no manual config file editing required.

Also supports manual MCP configuration — see the Integration Guide.

That's it. Next time you open a conversation, your AI will remember you.

How it works

Active Graph

All information — AI conversations, note content, your preferences and decisions — lives as nodes and links in a single graph. Nodes aren't static index cards; they carry four maturity dimensions (activity, refinement, connectivity, independence) that together determine a memory's fate: reinforced, connected, crystallized, or forgotten.

Learn more about the Active Graph model

Three Cognitive Tools

TideMind exposes three tools via the MCP protocol, corresponding to three cognitive operations:

Tool	What it does	When to use
brain_prepare	Load memory context	Start of every conversation
brain_recall	Navigate and retrieve memories	When you need historical context
brain_digest	Store new information	When something worth remembering emerges

This isn't CRUD. Prepare is like opening a memory map. Recall is like remembering — and each recall strengthens the memory. Digest is like absorption — incoming information is decomposed, connected, and woven into the existing knowledge network.

brain_recall goes beyond search. AI can follow links between memories — starting from an architecture decision, jumping to the discussion behind it, then to later changes it triggered. Like walking through a knowledge network, not guessing keywords in a search box. It returns a small set of highly relevant nodes first, then expands along links when more context is needed — exploring on demand, never loading everything upfront.

Learn more about the three tools

Metabolism

TideMind continuously maintains the graph in the background, much like the brain organizes memories during sleep:

• Daily: Inactive memories gradually decay (but highly connected hub nodes are protected); candidate links are confirmed or pruned
• Weekly: Divergent scanning — discovering node pairs that share neighbors but lack direct links, evaluating whether hidden connections exist; highly connected information crystallizes into Crystals — distilled representations of your thinking patterns, behavioral preferences, and core decisions

Information isn't just stored. It's digested, organized, connected, and sometimes forgotten. That's what "living" means.

Learn more about the metabolism system

Supported Integrations

AI Tools

Tool	Status	Notes
Claude Code	Supported	Native MCP support
Cursor	Supported	MCP configuration
Windsurf	Supported	MCP configuration
Codex	Supported	MCP configuration
Any MCP-compatible tool	Supported	Standard protocol, plug and play

Note Systems

Source	Status	Notes
Logseq	Supported	File watching, incremental sync, properties promoted to tags
Obsidian	Supported	Vault import, Canvas support, wikilink resolution
Apple Notes	Supported	Read-only sync, attachment text extraction
More coming soon...

Desktop Client

TideMind ships with a desktop client for browsing your knowledge graph, monitoring system metabolism, and tuning parameters. It's an inspection tool — for observing what your external brain is thinking, discovering, and forgetting.

The main interface of TideMind isn't this app. It's whatever AI tool you're already using.

Dashboard — real-time pulse of your second brain

Brain Explorer — visualize how your memories connect across domains

Memory detail — four-dimensional maturity model and evidence chains

Documentation

Document	Description
Design Philosophy	From Bush (1945) to Clark & Chalmers (1998) — the cognitive science behind TideMind
Architecture	Active Graph model, four maturity dimensions, link type system
Metabolism & Self-Evolution	Decay mechanics, divergent scanning, crystal emergence, three-tier learning
Integration Guide	Detailed setup for each AI tool and note system
API Reference	Full parameter documentation for prepare / recall / digest

Contributing

Contributions of all kinds are welcome. See CONTRIBUTING.md to get started.

License

MIT

---

Inspired by 80 years of cognitive science — from Bush's Memex to modern memory reconsolidation research.