Data Export

Classifyre lets you take your findings and assets out of the app and into the tools your team already lives in — Excel, Google Sheets, Power BI, or any spreadsheet. There are two ways to do it:

• Download a CSV — a one-time snapshot of exactly what you’re looking at right now. Best for sharing, archiving, or a quick pivot table.
• Connect a live feed — a link your spreadsheet refreshes on demand, so the numbers update themselves. Best for recurring reports and dashboards.

Both honor the same search and filters you’ve set on screen, so you only export the rows you actually want.

Download a CSV

Every major table has a Download CSV button:

• Findings — one row per finding (category, asset, source, severity, status, confidence, matched content, detection dates, and metadata).
• Assets — one row per asset-with-finding, so an asset that triggered five findings appears as five rows. Great for slicing findings by asset attributes.
• Scan assets (inside a scan run) — one row per processed asset, with its status, finding totals, and any error.

Connect a live feed (auto-refresh)

If you build the same report every week, a live connection saves you the repeat work. Your spreadsheet holds a link to Classifyre; you click Refresh and the latest data flows in.

Classifyre exposes a query link for each dataset that returns data in pages. Your spreadsheet walks through the pages automatically, so even large datasets come through in full.

Choosing what comes through

Add filters to the link exactly like the on-screen filters. A few examples:

• High and critical findings only: …/search/findings/query?severity=HIGH&severity=CRITICAL
• Open PII findings for one source: …/search/findings/query?detectorType=PII&status=OPEN&sourceId=<source-id>
• Critical findings grouped by asset: …/search/assets/query?finding_severity=CRITICAL

Repeat a parameter to select several values (e.g. severity=HIGH&severity=CRITICAL). You can also set limit (rows per page, up to 10,000; default 1,000).

Set it up

let
    BaseUrl = "https://<your-classifyre-host>/api/search/findings/query",
    Query   = "limit=5000&severity=HIGH&severity=CRITICAL",

    // Fetch one page; cursor = null for the first page
    GetPage = (cursor as nullable text) =>
        let
            Url  = BaseUrl & "?" & Query &
                   (if cursor <> null then "&cursor=" & cursor else ""),
            Resp = Json.Document(Web.Contents(Url))
        in
            Resp,

    // Keep fetching while the response hands back a nextCursor
    Pages = List.Generate(
        () => GetPage(null),
        each _ <> null,
        each if [nextCursor] <> null then GetPage([nextCursor]) else null
    ),

    Rows  = List.Combine(List.Transform(Pages, each [items])),
    Result = Table.FromRecords(Rows)
in
    Result

=IMPORTDATA("https://<your-classifyre-host>/api/search/findings/export?severity=CRITICAL")

How paging works

Each response looks like this:

{
  "items": [ /* up to `limit` rows */ ],
  "nextCursor": "eyJ..."   // null when you've reached the last page
}

Keep requesting with ?cursor=<nextCursor> until nextCursor is null. The order stays stable between pages, so you never miss or double-count a row — sort and filter inside your spreadsheet once everything has loaded.

Which should I use?

Both always reflect the filters you choose — so whichever you pick, you’re working with exactly the slice of findings or assets you care about.