API overview

Dynition exposes a single GraphQL API. Anything you can do in the app — read purchase orders and the orders they become, manage companies and contacts, upload a PO, trigger a sync — you can do programmatically with the same API the app itself uses.

Endpoint

There is one endpoint per environment. Send a POST request with your query in the request body and your API token in the Authorization header.

# Production
https://api.dynition.com/admin/graphql
 
# Staging
https://api.stg.dynition.com/admin/graphql

You don't need to know which region your account lives in — requests are routed to the right place automatically.

Your first request

Authenticate with a Bearer token (see Authentication) and send a query. This one lists your most recent purchase orders:

curl https://api.dynition.com/admin/graphql \
  -H "Authorization: Bearer lnqat_your_token_here" \
  -H "Content-Type: application/json" \
  -d '{"query": "{ purchaseOrders { list(first: 5) { edges { node { id poNumber status } } } } }"}'

A successful response wraps your data under a top-level data key:

{
  "data": {
    "purchaseOrders": {
      "list": {
        "edges": [
          { "node": { "id": "...", "poNumber": "PO-1042", "status": "NEEDS_REVIEW" } }
        ]
      }
    }
  }
}

How it's organized

Top-level operations are grouped by domain. You select a group, then a field on it — for example purchaseOrders.list, purchaseOrders.purchaseOrder(id:), orders.list, companies.list, or files.createFileUpload.

Try it in the browser

Opening the endpoint in a browser serves GraphiQL, an interactive playground with schema-aware autocomplete and inline docs — a fast way to shape a query before you write any code:

https://api.dynition.com/admin/graphql

If you're signed in to Dynition in the same browser, it's already authenticated. Otherwise open the Headers tab and add your token:

{ "Authorization": "Bearer lnqat_your_token_here" }

Next steps

• Authentication — create a token and make authenticated requests.
• Pagination — page through large result sets.
• Errors — how errors are returned and how to handle them.