Best for
- Developers integrating QuickBooks Online data or accounting operations
- Local integrations using Claude Code or another MCP-compatible client
- Teams needing sandbox testing or production company access
Broad coverage with 145 tools, 29 entity types, and 11 reports; installation depends on Intuit app registration and OAuth, and the conflicting license statements should be verified.
Strongest · Documentation 8/20 Weakest · Reliability 4/20
Static review · not runListed 2026-08-14
Read the FMRS scoring method →Runtime:Node.js
QUICKBOOKS_CLIENT_ID
Client ID of the app you create on the Intuit Developer Portal; required, not secret.
QUICKBOOKS_CLIENT_SECRET
Client Secret from the app's Keys & Credentials page; required, secret.
QUICKBOOKS_REFRESH_TOKEN
Obtained by completing the OAuth handshake via npm run auth; required, secret.
QUICKBOOKS_REALM_ID
Realm ID of the target QuickBooks company (sandbox or production); required, not secret.
QUICKBOOKS_ENVIRONMENT
Set to sandbox or production to choose the environment.
QUICKBOOKS_DISABLE_WRITE
Set to true to suppress create_* tools; optional.
QUICKBOOKS_DISABLE_UPDATE
Set to true to suppress update_* tools; optional.
QUICKBOOKS_DISABLE_DELETE
Set to true to suppress delete_* tools; optional.
QUICKBOOKS_TOKEN_STORE_PATH
Absolute writable token-store path for read-only filesystems or per-tenant isolation; must be set in the host process env, not in .env.
QBO_PDF_OUTPUT_DIR
When set, get_invoice_pdf writes PDFs to this directory instead of returning inline base64.
Clone the repository, run npm install and npm run build. Copy .env.example to .env and provide the QuickBooks OAuth client credentials, refresh token, Realm ID, and environment. Register an app in the Intuit Developer Portal and complete the browser-based OAuth flow; sandbox supports a localhost callback, while production requires a public HTTPS callback. Configure the client to run dist/index.js as a node stdio subprocess.
{"mcpServers":{"quickbooks":{"command":"node","args":["path/to/mcp-quickbooks-online/dist/index.js"],"env":{"QUICKBOOKS_CLIENT_ID":"your_client_id","QUICKBOOKS_CLIENT_SECRET":"your_client_secret","QUICKBOOKS_REFRESH_TOKEN":"your_refresh_token","QUICKBOOKS_REALM_ID":"your_realm_id","QUICKBOOKS_ENVIRONMENT":"sandbox","QUICKBOOKS_DISABLE_WRITE":"false","QUICKBOOKS_DISABLE_UPDATE":"false","QUICKBOOKS_DISABLE_DELETE":"false"}}}}
Shown for Claude Code. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.
{
"servers": {
"quickbooks": {
"command": "node",
"args": [
"path/to/mcp-quickbooks-online/dist/index.js"
],
"env": {
"QUICKBOOKS_CLIENT_ID": "your_client_id",
"QUICKBOOKS_CLIENT_SECRET": "your_client_secret",
"QUICKBOOKS_REFRESH_TOKEN": "your_refresh_token",
"QUICKBOOKS_REALM_ID": "your_realm_id",
"QUICKBOOKS_ENVIRONMENT": "sandbox",
"QUICKBOOKS_DISABLE_WRITE": "false",
"QUICKBOOKS_DISABLE_UPDATE": "false",
"QUICKBOOKS_DISABLE_DELETE": "false"
}
}
}
}
Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).
claude mcp add quickbooks -e QUICKBOOKS_CLIENT_ID=your_client_id -e QUICKBOOKS_CLIENT_SECRET=your_client_secret -e QUICKBOOKS_REFRESH_TOKEN=your_refresh_token -e QUICKBOOKS_REALM_ID=your_realm_id -e QUICKBOOKS_ENVIRONMENT=sandbox -e QUICKBOOKS_DISABLE_WRITE=false -e QUICKBOOKS_DISABLE_UPDATE=false -e QUICKBOOKS_DISABLE_DELETE=false -- node path/to/mcp-quickbooks-online/dist/index.js
Run it in a terminal; replace any <…> placeholders with your own values first.
After starting, check that tools like get_invoice, search_customers and get_profit_and_loss appear in the client's tool list; then call search_customers — returning data from your QuickBooks sandbox company confirms the connection works.
Once connected, you can ask your AI assistant things like:
No matching tools
Listed from the project's documentation, not tested by this site.
QuickBooks Online MCP Server exposes QuickBooks Online functionality to AI assistants through the Model Context Protocol. It provides CRUD and search capabilities for 29 entity types, 11 financial reports, OAuth 2.0 authentication, and local stdio execution.
Manage your Hevy workout data from AI assistants via MCP.
Firecrawl's official MCP server for web search, scraping, and structured extraction for AI agents
One binary. One grammar. Evidence from the biomedical sources you already trust.
Encrypted, fully offline long-term memory for AI agents
Source revision 31a1dd5c3e17 Data synced 2026-10-11 Read the FMRS scoring method