Getting Started

Panterra is an AI-native unified API platform spanning 13 verticals and 42 providers. Integrate once and connect to accounting software, CRMs, payment processors, e-commerce platforms, HRIS tools, ticketing systems, and more — all through a single, consistent REST API.

Every response uses a normalized schema so your application logic stays the same regardless of which provider your customer uses. Panterra also provides built-in anomaly detection, an AI query engine, a real-time sync layer, and an MCP server for AI agent access.

Quick start

1. Sign up at the Panterra Dashboard and create an organization.
2. Generate an API key from the dashboard. Keys start with pt_.
3. Embed the Connect Widget in your app so your users can authorize their software systems.
4. Make your first API call using the connection ID returned by the widget.

Base URL

https://api.panterra.io/v1

All endpoints are prefixed with /v1. Vertical data endpoints additionally require a x-connection-id header identifying which connected account to query.

URL structure

Panterra organizes endpoints by vertical. Every vertical follows the same pattern:

/v1/{vertical}/{entity}
/v1/{vertical}/{entity}/:id

/v1/accounting/invoices
/v1/crm/contacts
/v1/payments/transactions
/v1/ecommerce/orders
/v1/hris/employees

Your first API call

const res = await fetch('https://api.panterra.io/v1/accounting/invoices', {
  headers: {
    'X-API-Key': 'pt_your_api_key',
    'x-connection-id': 'your_connection_id',
  },
});
const { data } = await res.json();
console.log(data); // Array of normalized invoice objects

Authentication

Panterra supports two authentication methods. You can use either one depending on your integration approach.

API Key authentication

The simplest method. Include your API key in the X-API-Key header on every request. API keys are scoped to your organization and can be created from the dashboard or via the API.

X-API-Key: pt_your_api_key

JWT authentication

For session-based flows, sign in with email and password to receive a JSON Web Token. Pass it in the Authorization header.

Authorization: Bearer <token>

Sign up

Creates a new user and organization. Returns the user object, organization, and a JWT token.

{
  "email": "dev@acme.com",
  "password": "a-strong-password",
  "name": "Jane Doe",
  "orgName": "Acme Corp"
}

{
  "user": { "id": "usr_...", "email": "dev@acme.com", "name": "Jane Doe" },
  "organization": { "id": "org_...", "name": "Acme Corp" },
  "token": "eyJhbGciOiJIUzI1NiIs..."
}

Log in

Authenticates an existing user. Returns the same shape as sign up.

{
  "email": "dev@acme.com",
  "password": "a-strong-password"
}

Create API key

Requires authentication (JWT or existing API key). Creates a new API key for your organization. The full key value is returned only once in the response—store it securely.

Get current user

Returns the authenticated user and their organization details.

Connections

A connection represents the authorized link between your Panterra organization and one of your customer's software systems. Each connection stores OAuth tokens, provider metadata, and the vertical it belongs to. You reference connections by their ID in the x-connection-id header for all data requests.

Supported providers

Panterra supports 42 providers across 13 verticals.

OAuth flow

To initiate a connection, redirect your user (or open a popup) to the provider's connect URL:

Replace {vertical} with the vertical slug (e.g. accounting, crm) and {provider} with the provider slug (e.g. quickbooks, hubspot). Panterra handles the full OAuth dance and redirects back to your configured callback URL with the new connection ID. The Connect Widget wraps this flow in a drop-in UI.

List connections

Returns all active connections for your organization.

{
  "data": [
    {
      "id": "conn_abc123",
      "provider": "quickbooks",
      "vertical": "accounting",
      "status": "active",
      "companyName": "Acme Corp",
      "createdAt": "2026-01-15T08:30:00Z"
    },
    {
      "id": "conn_def456",
      "provider": "hubspot",
      "vertical": "crm",
      "status": "active",
      "companyName": "Acme Corp",
      "createdAt": "2026-02-10T14:00:00Z"
    }
  ]
}

Delete a connection

Revokes the OAuth tokens and removes the connection. Cached data for this connection is purged. This action cannot be undone.

Connect Widget

The Connect Widget is a drop-in JavaScript component that handles the entire OAuth flow in a branded modal overlay. Your users pick their vertical and provider, authorize access in a popup window, and you receive the connection ID in a callback.

Installation

Add the script tag to your page:

<script src="https://panterra.io/connect.js"></script>

Usage

Panterra.connect({
  apiKey: 'pt_your_api_key',
  verticals: ['accounting', 'crm'], // optional: limit visible verticals
  providers: ['quickbooks', 'hubspot'], // optional: limit visible providers
  onSuccess: (connection) => {
    console.log('Connected:', connection.id, connection.provider, connection.vertical);
    // Store connection.id for future API calls
  },
  onError: (error) => {
    console.error('Connection failed:', error.message);
  },
  onClose: () => {
    console.log('Widget closed');
  },
});

Options

What happens behind the scenes

1. The widget opens a modal overlay with the Panterra-branded vertical and provider picker.
2. When the user selects a provider, a popup window navigates to the OAuth authorization URL.
3. After the user authorizes, the popup closes and the modal receives the new connection details.
4. Your onSuccess callback fires with the connection object containing id, provider, and vertical.

Provider Capabilities

Not every provider supports every entity type in its vertical. Panterra's capability system lets you discover what each provider supports before making API calls.

List all providers

Returns every supported provider grouped by vertical, including entity counts and auth types.

{
  "data": [
    {
      "slug": "quickbooks",
      "name": "QuickBooks Online",
      "vertical": "accounting",
      "authType": "oauth2",
      "supportedEntities": ["invoices", "contacts", "payments", "items", "accounts", "bills", "expenses", "credit_notes", "bank_transactions", "purchase_orders", "journal_entries"]
    },
    {
      "slug": "hubspot",
      "name": "HubSpot",
      "vertical": "crm",
      "authType": "oauth2",
      "supportedEntities": ["contacts", "companies", "deals", "activities", "pipelines", "notes"]
    }
  ]
}

Get provider capabilities

Returns detailed per-entity capability information for a specific provider.

{
  "provider": "quickbooks",
  "vertical": "accounting",
  "capabilities": {
    "invoices": { "list": true, "get": true, "create": true, "update": true, "delete": false },
    "contacts": { "list": true, "get": true, "create": true, "update": true, "delete": false },
    "payments": { "list": true, "get": true, "create": true, "update": false, "delete": false },
    "journal_entries": { "list": true, "get": true, "create": true, "update": false, "delete": false }
  }
}

Handling unsupported entities

If you call an endpoint for an entity that the connected provider does not support, Panterra returns a 501 Not Implemented response with a clear error message.

{
  "statusCode": 501,
  "error": "Not Implemented",
  "message": "The provider 'wave' does not support 'purchase_orders'. Use GET /v1/providers/wave/capabilities to see supported entities."
}

Accounting

The accounting vertical provides a normalized interface to financial data from QuickBooks, Xero, FreshBooks, Wave, Sage, and NetSuite. Create, read, and update invoices, contacts, payments, and more through a single API.

Invoices

Query parameters

const res = await fetch('https://api.panterra.io/v1/accounting/invoices?limit=20', {
  headers: {
    'X-API-Key': apiKey,
    'x-connection-id': connectionId,
  },
});
const { data, cursor } = await res.json();

{
  "contactId": "ct_abc123",
  "dueDate": "2026-04-15",
  "lineItems": [
    {
      "description": "Consulting services",
      "quantity": 10,
      "unitAmount": 150.00,
      "accountId": "acc_revenue_01"
    }
  ],
  "memo": "March consulting engagement"
}

Invoice object fields

Contacts

const res = await fetch('https://api.panterra.io/v1/accounting/contacts', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Contact object fields

Payments

{
  "contactId": "ct_abc123",
  "date": "2026-03-15",
  "amount": 1500.00,
  "currency": "USD",
  "accountId": "acc_checking_01",
  "allocations": [
    { "invoiceId": "inv_xyz789", "amount": 1500.00 }
  ],
  "reference": "CHK-4521"
}

Payment object fields

Additional accounting endpoints

The following entities follow the same list/get/create/update pattern.

Items

Accounts (Chart of Accounts)

Bills

Expenses

Credit Notes

Bank Transactions

Purchase Orders

Journal Entries

CRM

The CRM vertical provides unified access to customer relationship data across Salesforce, HubSpot, and Pipedrive. Manage contacts, companies, deals, activities, and pipelines through a single interface.

Contacts

const res = await fetch('https://api.panterra.io/v1/crm/contacts', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

CRM Contact object fields

Deals

{
  "id": "deal_abc123",
  "remoteId": "9012345",
  "name": "Acme Corp Enterprise",
  "amount": 85000.00,
  "currency": "USD",
  "stage": "Proposal Sent",
  "pipelineId": "pipe_main",
  "contactId": "ct_xyz789",
  "companyId": "comp_acme",
  "probability": 60,
  "expectedCloseDate": "2026-04-30",
  "createdAt": "2026-02-01T10:00:00Z",
  "updatedAt": "2026-03-15T14:30:00Z"
}

Additional CRM endpoints

Companies

Activities

Pipelines

Notes

Leads

Payments

The payments vertical provides unified access to payment processing data from Stripe, Square, and PayPal. Read transactions, manage customers, process refunds, and track payouts.

Transactions

const res = await fetch('https://api.panterra.io/v1/payments/transactions', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Transaction object fields

Customers

Customer object fields

Additional payments endpoints

Refunds

Payouts

Subscriptions

Payment Methods

Disputes

Balance

E-commerce

The e-commerce vertical connects to online stores and marketplaces via Shopify, WooCommerce, and BigCommerce. Manage products, orders, customers, collections, fulfillments, inventory, and discounts.

Products

const res = await fetch('https://api.panterra.io/v1/ecommerce/products', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Product object fields

Orders

{
  "id": "ord_abc123",
  "remoteId": "1001",
  "orderNumber": "#1001",
  "status": "FULFILLED",
  "financialStatus": "PAID",
  "customer": { "id": "cust_xyz", "email": "jane@acme.com" },
  "totalAmount": 299.00,
  "currency": "USD",
  "lineItems": [
    { "productId": "prod_abc", "title": "Widget Pro", "quantity": 2, "price": 149.50 }
  ],
  "shippingAddress": { "city": "Austin", "state": "TX", "country": "US" },
  "createdAt": "2026-03-15T10:30:00Z"
}

Additional e-commerce endpoints

Customers

Collections

Fulfillments

Inventory

Discounts

HRIS

The HRIS vertical provides unified access to human resources data from BambooHR, Gusto, and Rippling. Manage employee records, departments, time off, compensation, payroll, benefits, and office locations.

Employees

const res = await fetch('https://api.panterra.io/v1/hris/employees', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Employee object fields

Additional HRIS endpoints

Departments

Time Off

Compensation

Payroll

Benefits

Locations

Ticketing

The ticketing vertical connects to project management and issue tracking systems including Jira, Asana, and Linear. Manage projects, tickets, comments, sprints, and attachments.

Tickets

const res = await fetch('https://api.panterra.io/v1/ticketing/tickets', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Ticket object fields

Additional ticketing endpoints

Projects

Comments

Users

Sprints

Attachments

Messaging

The messaging vertical connects to team communication platforms including Slack, Microsoft Teams, and Discord. Read channels, messages, and user information.

Channels

const res = await fetch('https://api.panterra.io/v1/messaging/channels', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Channel object fields

Messages

Users

Support

The support vertical connects to customer support platforms including Zendesk, Intercom, and Freshdesk. Manage support tickets, contacts, conversations, and knowledge base articles.

Tickets

const res = await fetch('https://api.panterra.io/v1/support/tickets', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Support Ticket object fields

Additional support endpoints

Contacts

Conversations

Articles (Knowledge Base)

File Storage

The file storage vertical connects to cloud storage platforms including Google Drive, Dropbox, and Box. Browse, read, and manage files and folders.

Files

const res = await fetch('https://api.panterra.io/v1/filestorage/files?folderId=folder_root', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

File object fields

Folders

Analytics

The analytics vertical connects to product analytics and web analytics platforms including Google Analytics, Mixpanel, and Amplitude. Pull reports and event data.

Reports

const res = await fetch('https://api.panterra.io/v1/analytics/reports?startDate=2026-03-01&endDate=2026-03-18', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Events

Email

The email vertical connects to email platforms including Gmail, Outlook, and SendGrid. Read and send messages, manage threads, and work with contact lists.

Messages

const res = await fetch('https://api.panterra.io/v1/email/messages?limit=20', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Email Message object fields

Threads

Contact Lists

Scheduling

The scheduling vertical connects to appointment and calendar tools including Calendly, Cal.com, and Google Calendar. Manage event types, bookings, and availability windows.

Event Types

const res = await fetch('https://api.panterra.io/v1/scheduling/event-types', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Event Type object fields

Bookings

Availability

ATS

The ATS (Applicant Tracking System) vertical connects to recruiting platforms including Greenhouse, Lever, and Workday. Manage candidates, jobs, applications, and interviews.

Candidates

const res = await fetch('https://api.panterra.io/v1/ats/candidates', {
  headers: { 'X-API-Key': apiKey, 'x-connection-id': connectionId },
});
const { data } = await res.json();

Candidate object fields

Jobs

{
  "id": "job_abc123",
  "remoteId": "4567",
  "title": "Senior Backend Engineer",
  "departmentId": "dept_eng",
  "status": "OPEN",
  "hiringManagerId": "emp_jane",
  "location": "Remote",
  "employmentType": "FULL_TIME",
  "postedAt": "2026-02-15T09:00:00Z",
  "createdAt": "2026-02-10T08:00:00Z"
}

Additional ATS endpoints

Applications

Interviews

Anomaly Detection

Panterra provides a three-layer anomaly detection system that automatically identifies suspicious patterns in accounting data. Each layer adds increasing intelligence:

1. Deterministic layer — Rule-based checks for known red flags (duplicate invoices, missing fields, negative amounts, round-number patterns).
2. Statistical layer — Z-score and IQR-based outlier detection on amounts, frequencies, and timing patterns.
3. LLM layer — AI-powered contextual analysis that identifies subtle anomalies a rules engine would miss.

Run a scan

Query parameters

{
  "anomalies": [
    {
      "id": "anom_001",
      "type": "DUPLICATE_INVOICE",
      "severity": "critical",
      "entityType": "invoice",
      "entityId": "inv_abc123",
      "description": "Invoice #1042 appears to be a duplicate of #1041",
      "layer": "deterministic",
      "status": "open",
      "detectedAt": "2026-03-17T12:00:00Z"
    }
  ],
  "summary": {
    "total": 12,
    "critical": 2,
    "warning": 7,
    "info": 3
  }
}

List anomalies

Query parameters

Resolve an anomaly

Marks the anomaly as resolved. Use this after the underlying issue has been corrected.

Dismiss an anomaly

Marks the anomaly as dismissed (false positive or intentional).

Inline detection

You can enable anomaly detection on any accounting list endpoint by adding ?detect_anomalies=true as a query parameter. This runs the deterministic layer against the returned results and includes any findings in the response.

GET /v1/accounting/invoices?detect_anomalies=true

Anomaly types

AI Queries

The AI query endpoint lets you ask natural-language questions about your connected data. Panterra uses an LLM agent with access to your data across all connected verticals to generate answers, pulling real figures from your business systems.

Ask a question

Requires both authentication and the x-connection-id header.

{
  "question": "What are my top 5 overdue invoices?"
}

{
  "question": "What are my top 5 overdue invoices?",
  "answer": "Here are your top 5 overdue invoices by amount:\n\n1. INV-1089 — $12,500.00 (Acme Corp, 45 days overdue)\n2. INV-1076 — $8,200.00 (Global Supplies, 30 days overdue)\n...",
  "toolsUsed": ["list_invoices", "list_contacts"]
}

Example questions

• "What are my top 5 overdue invoices?"
• "Show me revenue this month"
• "Which customers have outstanding balances over $10,000?"
• "Compare this quarter's expenses to last quarter"
• "Are there any duplicate invoices?"
• "Summarize payments received this week"
• "How many open deals are in the pipeline?"
• "What is our employee headcount by department?"

Sync Engine

Panterra maintains a local cache of your data for fast reads. Instead of hitting the provider's API on every request (which can take 300-500ms), cached reads return in approximately 10ms. The sync engine keeps this cache fresh through multiple strategies.

How sync works

• Background sync — Every 5 minutes, Panterra pulls incremental updates from the provider for each active connection.
• Webhook-triggered — When a provider supports outbound webhooks, changes are synced in near real-time.
• Write-through — When you create or update data through Panterra, the write goes directly to the provider and the cache is updated immediately.
• On-demand — Add ?fresh=true to any read endpoint to bypass the cache and fetch directly from the provider.

View sync status

Returns the sync state for each entity type per connection.

{
  "connectionId": "conn_abc123",
  "vertical": "accounting",
  "entities": {
    "invoices": {
      "lastSyncAt": "2026-03-17T11:55:00Z",
      "status": "synced",
      "recordCount": 1247
    },
    "contacts": {
      "lastSyncAt": "2026-03-17T11:55:00Z",
      "status": "synced",
      "recordCount": 342
    },
    "payments": {
      "lastSyncAt": "2026-03-17T11:55:00Z",
      "status": "syncing",
      "recordCount": 891
    }
  }
}

Trigger a manual sync

Requires the x-connection-id header. Starts an immediate sync for all entity types on the specified connection. Returns 202 Accepted as the sync runs asynchronously.

Webhooks

Subscribe to real-time notifications when data changes in connected systems. Panterra sends HTTP POST requests to your endpoint with event payloads.

Create a webhook subscription

{
  "url": "https://your-app.com/webhooks/panterra",
  "events": ["accounting.invoice.*", "crm.deal.updated", "payments.transaction.created"]
}

{
  "id": "wh_abc123",
  "url": "https://your-app.com/webhooks/panterra",
  "events": ["accounting.invoice.*", "crm.deal.updated", "payments.transaction.created"],
  "secret": "whsec_a1b2c3d4e5f6...",
  "active": true,
  "createdAt": "2026-03-17T12:00:00Z"
}

List webhook subscriptions

View delivery history

Returns a list of recent delivery attempts with status codes and timestamps.

Delete a webhook

Event patterns

Events follow the pattern {vertical}.{entity}.{action}.

Signature verification

Every webhook delivery includes an X-Panterra-Signature header containing an HMAC-SHA256 signature of the request body using your webhook secret. Always verify this signature before processing the payload.

import { createHmac } from 'crypto';

function verifyWebhook(body: string, signature: string, secret: string): boolean {
  const expected = createHmac('sha256', secret)
    .update(body)
    .digest('hex');
  return signature === expected;
}

Retry policy

If your endpoint returns a non-2xx status code or fails to respond within 10 seconds, Panterra retries with exponential backoff:

After 5 failed retries the delivery is marked as failed. Check the delivery history endpoint to diagnose issues.

MCP Server

Panterra includes a Model Context Protocol (MCP) server, enabling AI assistants like Claude to interact directly with your business data across all verticals. The server exposes tools that an LLM can call to read and write data.

Transport options

Panterra supports two MCP transport modes:

Stdio (local)

Run the MCP server as a local process that communicates over stdin/stdout. Ideal for Claude Desktop and other local AI tools.

PANTERRA_API_KEY=pt_xxx node dist/mcp/start-mcp.js

HTTP (remote)

Send MCP protocol messages over HTTP for remote or server-side integrations.

Include your API key via the x-api-key header.

Available tools

Read-only mode

For safety, you can restrict the MCP server to read-only operations by setting the environment variable:

PANTERRA_MCP_MODE=readonly

In read-only mode, all write tools (create_*) are disabled.

Claude Desktop configuration

Add the following to your Claude Desktop claude_desktop_config.json file to connect Panterra as an MCP server:

{
  "mcpServers": {
    "panterra": {
      "command": "node",
      "args": ["path/to/panterra-api/dist/mcp/start-mcp.js"],
      "env": {
        "PANTERRA_API_KEY": "pt_your_api_key"
      }
    }
  }
}

SDKs

Official client libraries make it easy to integrate Panterra into your application. Both SDKs provide typed interfaces, automatic pagination, and error handling across all 13 verticals.

TypeScript / JavaScript

npm install @panterra/sdk

import { PanterraClient } from '@panterra/sdk';

const client = new PanterraClient({ apiKey: 'pt_your_api_key' });

// Accounting: List invoices
const invoices = await client.accounting.invoices.list('conn_abc123');

// CRM: List deals
const deals = await client.crm.deals.list('conn_def456');

// E-commerce: List orders
const orders = await client.ecommerce.orders.list('conn_ghi789');

// Auto-paginate through all invoices
for await (const inv of client.accounting.invoices.listAll('conn_abc123')) {
  console.log(inv.invoiceNumber, inv.totalAmount);
}

// Create an invoice
const newInvoice = await client.accounting.invoices.create('conn_abc123', {
  contactId: 'ct_abc123',
  dueDate: '2026-04-15',
  lineItems: [
    { description: 'Consulting', quantity: 10, unitAmount: 150 },
  ],
});

Python

pip install panterra

from panterra import PanterraClient

client = PanterraClient(api_key="pt_your_api_key")

# Accounting: List invoices
invoices = client.accounting.invoices.list("conn_abc123")

# CRM: List deals
deals = client.crm.deals.list("conn_def456")

# E-commerce: List orders
orders = client.ecommerce.orders.list("conn_ghi789")

# Auto-paginate through all invoices
for inv in client.accounting.invoices.list_all("conn_abc123"):
    print(inv.invoice_number, inv.total_amount)

# Create an invoice
new_invoice = client.accounting.invoices.create(
    "conn_abc123",
    contact_id="ct_abc123",
    due_date="2026-04-15",
    line_items=[
        {"description": "Consulting", "quantity": 10, "unit_amount": 150},
    ],
)

API Reference

For the complete auto-generated API reference with request/response schemas for all 13 verticals, visit the interactive API Reference page powered by Swagger/OpenAPI.