Paginate MCP tool results

the platform list tools return one page at a time. If a tool’s input schema includes size and cursor, continue through its results by passing the returned next_cursor into the next call.

Pagination loop

1. Call the list tool without cursor.
2. Read the items from the tool-specific result field.
3. If the result includes next_cursor, call the same tool again with that value as cursor.
4. Stop when next_cursor is absent.

Treat cursor values as opaque. Do not decode, edit, or construct them.

Example with listRisks

The first call includes the organization and, optionally, a page size:

{
  "organization_id": "organization-id",
  "size": 50
}

listRisks returns its items in risks. When another page is available, the result also includes next_cursor:

{
  "risks": [
    {
      "id": "risk-id",
      "name": "Unauthorized production access"
    }
  ],
  "next_cursor": "cursor-value"
}

Use that value in the next call:

{
  "organization_id": "organization-id",
  "size": 50,
  "cursor": "cursor-value"
}

Continue until the result no longer includes next_cursor.

Keep the query unchanged

A cursor belongs to the query that produced it. Keep the organization, filters, ordering, and page size unchanged when requesting the next page:

{
  "organization_id": "organization-id",
  "size": 50,
  "filter": {
    "query": "access"
  },
  "cursor": "cursor-value"
}

Filtering and ordering options differ by tool. Inspect the tool’s input schema in your MCP client instead of assuming that every list tool accepts the same fields.

Result fields differ by tool

The collection field matches the resource being listed. For example, listRisks returns risks, while listThirdParties returns thirdParties. Use the output schema advertised by the server to determine the field for a specific tool.

For long-running jobs, process each page before requesting the next one instead of accumulating the complete result set in memory.