NodeSpace

Faster context. Fewer tokens.

AI coding assistants forget everything between sessions. NodeSpace gives them persistent, searchable access to your project knowledge — so you stop re-explaining your codebase every time you start a conversation.

nodespace.ai · Download

⚠️ Alpha Preview — NodeSpace is in early development. Features may change and data formats are not yet stable.

Why NodeSpace

Developers using AI assistants waste time copying files, re-explaining architecture, and watching context degrade mid-session. NodeSpace fixes this by sitting between your knowledge and your AI tools:

• 80% fewer roundtrips — AI agents query your knowledge base via MCP instead of scanning files with grep/ripgrep
• Runs entirely on your machine — no cloud accounts, no API calls, no data leaving localhost
• Works offline — on planes, behind VPNs, anywhere

You write things down once. Every AI tool you use can find them instantly.

Features

📅 Daily Journal

Quick access to today's context. One click opens a new entry for the current date, making it easy to capture thoughts throughout the day.

📝 Hierarchical Nodes

Create nested, indented blocks of content. Organize complex ideas with unlimited depth — like Logseq or Roam Research.

🏷️ Collections

Flexible organization that combines the best of folders and tags. A single node can belong to multiple collections without duplicating content.

🔗 @Mentions & Linking

Type @ to link to any node. Build a knowledge graph by connecting related ideas. A backlinks panel shows everything that references each node.

✅ Task Management

Markdown-style tasks ([ ], [x], [~]) with visual checkboxes. Track progress on projects while keeping tasks connected to their context.

🔍 Semantic Search

Find what you mean, not just what you typed. Ask "Where do we handle authentication?" and find relevant content without exact keyword matches.

🤖 MCP Integration

Built-in MCP server for AI tools. Opens with the app — Claude Code, Cursor, Codex, and any MCP-compatible assistant can query your knowledge base locally.

Installation

Download the Desktop App

Download NodeSpace →

Build from Source

Prerequisites:

• Bun 1.0+ — curl -fsSL https://bun.sh/install | bash
• Rust 1.80+ — curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

git clone https://github.com/NodeSpaceAI/nodespace-core
cd nodespace-core
bun install
bun run tauri:dev

MCP Setup

NodeSpace includes a built-in MCP server that starts automatically when you open the app. Your AI tools connect to it locally — there's nothing to deploy or host.

Note: The MCP server binds to localhost only and is accessible to other processes on your machine. Authentication is planned for a future release.

Claude Code

Add to your Claude Code MCP settings (~/.claude.json or via Claude Code settings):

{
  "mcpServers": {
    "nodespace": {
      "type": "stdio",
      "command": "/path/to/nodespace-core/scripts/mcp-bridge.sh"
    }
  }
}

Cursor / Codex / Other MCP Clients

For tools that support HTTP transport directly:

{
  "mcpServers": {
    "nodespace": {
      "type": "http",
      "url": "http://localhost:3100/mcp"
    }
  }
}

Verify Connection

With NodeSpace running, test the connection:

curl -X POST http://localhost:3100/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'

Semantic Search

Once connected via MCP, your AI assistant can query your knowledge base semantically.

Example Queries

Ask your AI assistant questions like:

• "What is our development process for picking up issues?"
• "How do we handle authentication in the backend?"
• "What are the coding standards for this project?"
• "Find context related to the quarterly planning meeting"

Using the search_semantic Tool

AI tools can call the semantic search directly:

{
  "name": "search_semantic",
  "arguments": {
    "query": "development process for implementing issues",
    "limit": 5
  }
}

This returns relevant nodes ranked by semantic similarity — not just keyword matches.

Filtering by Collection

Narrow searches to specific areas:

{
  "name": "search_semantic",
  "arguments": {
    "query": "validation flow",
    "collection": "Architecture"
  }
}

Quick Start

1. Open Daily Journal — Click "Daily Journal" in the sidebar to start today's entry
2. Create content — Just start typing. Press Enter to create a new block below
3. Organize with nesting — Press Tab to indent a block under the one above. Press Shift+Tab to outdent
4. Link your knowledge — Type @ to search and link to any other node
5. Use Collections — Expand "Collections" in the sidebar to organize content into categories
6. Connect your AI — Configure MCP (see above) and ask your AI assistant to search your knowledge base

Roadmap

See the open issues for the full backlog.

Contributing

NodeSpace is built with Rust, Svelte 5, SurrealDB, and Tauri.

bun run dev          # Browser development mode
bun run tauri:dev    # Desktop app development
bun run test         # Run tests
bun run build        # Production build

• Architecture docs: docs/architecture/
• AI agent dev guide: CLAUDE.md — conventions and workflow for developing with AI assistants

We welcome contributions. If you're thinking about a larger change, open an issue first so we can discuss the approach.

Community

• 🌟 Star this repo if NodeSpace is useful to you
• 🐛 Report a bug

License

NodeSpace is licensed under the Functional Source License 1.1 (Apache 2.0).

• ✅ Use NodeSpace freely for any purpose
• ✅ Modify the code to fit your needs

See LICENSE for the full text.