Desktop Setup (Claude Desktop)

Run Crow locally with Claude Desktop using stdio transport. No cloud deployment needed — everything runs on your machine.

Prerequisites

• Claude Desktop installed
• Node.js 18 or later
• Git

Step 1: Clone and Install

git clone https://github.com/kh0pper/crow.git
cd crow
npm run setup

This installs dependencies and creates the local SQLite database at data/crow.db.

Step 2: Configure API Keys

Copy the example env file and add your keys:

cp .env.example .env

Edit .env and add keys for the services you want. All core servers (memory, research, sharing, blog) work without any API keys — only storage (requires MinIO) and external integrations need keys.

Security note: Your .env file contains API keys that act like passwords. Never share it, commit it to GitHub, or post its contents anywhere. The file is already listed in .gitignore so git won't track it, but always double-check before pushing code. See the Security Guide for more about keeping your keys safe.

Step 3: Generate Claude Desktop Config

npm run desktop-config

This outputs a JSON block for your Claude Desktop configuration. Copy it and add it to:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

Step 4: Restart Claude Desktop

Quit and reopen Claude Desktop. You should see the MCP server icons in the input area, indicating Crow is connected.

Verify and Connect Your AI

Check that everything is working:

npm run check

Try it out — in Claude Desktop, say:

"Remember that today is my first day using Crow" "What do you remember?"

Reducing Context Usage

By default, each Crow server is a separate entry in your Claude Desktop config. For fewer context tokens, use npm run mcp-config -- --combined to generate a single crow-core entry that activates servers on demand. See the Context & Performance guide for details.

Limitations

• Only works with Claude Desktop (stdio transport)
• Only accessible from the machine running the servers
• No OAuth — direct process connection
• For mobile/web access, use the Cloud Deploy instead

Want remote access?

Chain this desktop instance with a cloud instance for remote access and sync. Set up Oracle Cloud or Google Cloud, then chain them together — your memories stay in sync across both.