Skip to content
Eternify Labs

Integrations

Connected to the tools you already use.

Bank feeds, payment processors, payroll, CRM, commerce and spend tools connect natively, with no middleware to pay for or maintain. Your data goes to your warehouse, and your agents work the books through MCP.

Integrations, in motion
  • Native, not middlewareEach connector is built into the ledger and runs inside its approvals. There is no third-party automation service in the middle to break or bill you.
  • Data lands as proposalsBank lines, payouts and receipts arrive as proposed entries with their source attached. A named person approves what posts.
  • Open at both endsImport from the system you leave, export to the warehouse you keep, and let agents in through the MCP connector with the same roles as people.

What you get

Native, not middleware

Each connector is built into the ledger and runs inside its approvals. There is no third-party automation service in the middle to break or bill you.

Data lands as proposals

Bank lines, payouts and receipts arrive as proposed entries with their source attached. A named person approves what posts.

Open at both ends

Import from the system you leave, export to the warehouse you keep, and let agents in through the MCP connector with the same roles as people.

How it works

Connect it once, reconcile continuously.

Pick a step to see it. The panel stays beside you while you read.

Connect an account through Plaid and the lines arrive daily and are matched to invoices and bills as they land. A bank without a feed uploads a CSV statement, mapped once and reused.

Bank feed

  1. Account connectedPlaid, read-onlyDone
  2. Lines arriveDailyDone
  3. Matched to the books212 of 219 linesDone
  4. Exceptions for a person7 linesWaiting

Stripe, Moneris, Square, Shopify, Amazon and Lightspeed payouts are broken into gross sales, refunds, fees and the net deposit, so the bank line reconciles to the cent.

Stripe payout

Gross charges
5,000.00
Processing fees
-145.00
Net deposit
4,855.00

ADP, Wagepoint, Payworks, Humi and Dayforce pay runs come in as a general ledger file. Accounts are mapped once, and each run is proposed as a journal for approval.

Payroll journal

Pay run, 15 OctoberGross wages and employer costs84,300.00Awaiting approver
Pay run, 30 SeptemberGross wages and employer costs83,950.00Posted

HubSpot and Salesforce bring customers and closed deals into receivables and billing. Ramp and Dext bring card transactions and receipts with fields already read.

Arrived today

Closed deal, Northgate FoodsFrom the CRM, contract attached30,000.00Review
Card transactionFrom the spend tool, receipt attached186.40Review
Supplier billRead from a forwarded email4,812.00Coded

Import the chart of accounts, balances, customers and vendors from QuickBooks Online, Xero or Sage 50, or from any CSV with a column mapping and a preview before anything posts.

Import

  1. Choose the sourceQuickBooks Online, Xero, Sage 50, CSV or PDFInfo
  2. Map the columnsPreview before postingInfo
  3. Opening balances reviewedTrial balance agrees to the sourceWaiting
  4. PostedApproved by a named personDone

Send ledger tables to Snowflake or BigQuery on a schedule, or export reports to Excel and Google Sheets. Claude, ChatGPT, Microsoft Copilot Studio, Cursor or any MCP client can work the books with the role of the person who connected it.

Which Stripe payouts from last week are not matched to a bank line?

Read payouts, last week ok

Read bank lines, last week ok

Two payouts, 4,855.00 and 2,310.00, have no bank line yet. Both are due tomorrow.

Bank feed

  1. Account connectedPlaid, read-onlyDone
  2. Lines arriveDailyDone
  3. Matched to the books212 of 219 linesDone
  4. Exceptions for a person7 linesWaiting

Under the hood

Everything arrives as a proposal.

Every connector writes to the same intake, so a bank line, a payout and a receipt are reviewed the same way, under the same roles.

By category

Every connection, by name.

Native connectors, built and kept by us, never through middleware.

Banking

  • Plaid bank feedsDaily bank and card lines, matched on arrival
  • CSV statementsAny bank's export, mapped once and reused

Payments

  • StripeCharges, refunds, fees and payouts settled line by line
  • MonerisCard settlements matched to invoices
  • SquareSales and payouts posted to the ledger

Payroll

  • ADPPay run journal imported as a general ledger file
  • WagepointPay run journal imported as a general ledger file
  • PayworksPay run journal imported as a general ledger file
  • HumiPay run journal imported as a general ledger file
  • DayforcePay run journal imported as a general ledger file

CRM and sales

  • HubSpotCustomers and closed deals into receivables
  • SalesforceAccounts and contracts into billing

Commerce

  • ShopifyOrders, fees and payouts per settlement
  • AmazonSettlement reports posted with marketplace fees
  • LightspeedSales and stock movements into the ledger

Spend and receipts

  • RampCard transactions arrive coded for review
  • DextReceipts and bills arrive with fields read

Data

  • SnowflakeScheduled warehouse export of ledger tables
  • BigQueryScheduled warehouse export of ledger tables
  • Excel and Google SheetsReports exported with their drill references

Import from

  • QuickBooks OnlineChart of accounts, balances, customers and vendors
  • XeroChart of accounts, balances, customers and vendors
  • Sage 50Chart of accounts, balances, customers and vendors
  • CSVAny list, with a column mapping and a preview
  • PDFBills and statements kept as evidence and read

AI agents

  • ClaudeWorks the books through the MCP connector
  • ChatGPTWorks the books through the MCP connector
  • Microsoft Copilot StudioWorks the books through the MCP connector
  • CursorWorks the books through the MCP connector
  • Any MCP clientSame tools, same roles, same approvals

Product names are trademarks of their respective owners and are used only to describe compatibility. No endorsement or partnership is implied.

Keep going with the rest.

  • MCP connector

    Your books in Claude, ChatGPT or any agent that speaks MCP.

    Learn more
  • API and webhooks

    Build on your books.

    Learn more
  • Reconciliation

    Bank, card and Stripe matched continuously.

    Learn more

Questions

Questions, answered.

Does Eternify Ledger connect to Canadian banks?

Yes. Bank and card accounts connect through Plaid for daily feeds. A bank without a feed uploads a CSV statement, and the column mapping is saved for next time.

Can I move from QuickBooks Online, Xero or Sage 50?

Yes. Import the chart of accounts, balances, customers and vendors from any of the three, or from a CSV, with a preview before anything posts and a trial balance check against the source.

Which payroll providers does it work with?

ADP, Wagepoint, Payworks, Humi and Dayforce pay run files come in as a general ledger journal. Accounts are mapped once and each run is proposed for approval.

Do I need Zapier or another middleware?

No. The connectors are built into the ledger. For anything else, the API and webhooks, and the MCP connector for agents, are open to your own systems.

Can I send data to a warehouse?

Yes. Ledger tables export to Snowflake or BigQuery on a schedule, and reports export to Excel and Google Sheets.

Connect the tools you run on.

Tell us your bank, payroll and payment tools and we show them connected on a working ledger.