Claude Desktop
Dinobase integrates with Claude Desktop via MCP. Once connected, Claude can query all your synced business data through natural conversation.
Install
Section titled “Install”curl -fsSL https://dinobase.ai/install.sh | bash -s -- claude-desktopInstalls Dinobase via uv, runs dinobase init, and writes the mcpServers entry to your Claude Desktop config file (~/Library/Application Support/Claude/claude_desktop_config.json on macOS). Then connect your data sources:
dinobase add stripe --api-key sk_test_...dinobase add hubspot --api-key pat-...dinobase syncSee Connecting Sources for the full list of 100+ supported sources, and Syncing & Scheduling for background sync options.
To re-run the setup step: dinobase install claude-desktop (safe to run multiple times — merges rather than overwrites).
With background sync
Section titled “With background sync”Keep data fresh while Claude Desktop is running:
{ "mcpServers": { "dinobase": { "command": "dinobase", "args": ["serve", "--sync", "--sync-interval", "30m"] } }}How it works
Section titled “How it works”When Claude Desktop connects, it sees dynamic instructions computed from your actual database state — which sources are connected, what tables exist, and how many rows are loaded. Claude will:
- Call
list_sourcesto discover available data - Call
describeon relevant tables to understand columns and types - Write and execute SQL via
query - For mutations (UPDATE/INSERT),
queryreturns a preview — Claude callsconfirmto execute
Available tools
Section titled “Available tools”The MCP server exposes these tools to Claude:
| Tool | Description |
|---|---|
query | Execute SQL queries (DuckDB dialect) |
describe | Get table schema, column types, and sample data |
list_sources | List all connected sources with row counts and freshness |
refresh | Re-sync a data source to get fresh data |
confirm | Execute a pending mutation (write-back to source API) |
confirm_batch | Execute multiple pending mutations |
cancel | Cancel a pending mutation |
Next steps
Section titled “Next steps”- Getting Started — Full setup walkthrough
- Connecting Sources — Add your business data
- Querying Data — SQL patterns and cross-source joins
- Syncing & Scheduling — Keep data fresh
- Schema Annotations — Add context for AI agents
- MCP Integration — How the MCP server works
- MCP Tools Reference — Detailed tool schemas