Smart Paste Format Detector

Paste unfamiliar content to see a likely format and the evidence behind it. Open a suggested tool when you are ready to work with that format.

Runs locally in your browser
Detection runs entirely in this browser. Content is not uploaded, stored, or opened automatically.
Unknown payloadJSON, JWT, cURL, URL, CSV, XML, PEM, SQL, timestamps and more

Identify before transforming

Smart Paste uses conservative signatures and strict parsers where possible. Review the evidence before following a suggested workflow; detection is guidance, not a security verdict.

How to identify a pasted payload

This page reads a payload you paste and names the format it most likely is, with a confidence figure, the evidence behind the guess, and links to the VoriTools tools that work with that format. Everything runs in the browser tab; the text is not uploaded, stored or opened.

Detection is heuristic. A strict JSON parse, a PEM header or a UUID pattern is a real test; the other rules look for shapes, such as two lines of key: value or the same comma count on every row. When the signals are weak the page says so with a lower confidence and plain text instead of a guess.

  1. Paste the payload into the field on the left. The tool does not analyse as you type, so nothing changes until you press Detect format.
  2. Press Detect format. The detected format and its confidence appear in the right-hand card, followed by the evidence lines that produced the result.
  3. Read the evidence. Strict parsers (JSON, PEM, UUID) give a definitive label, while CSV, YAML, HTML and Base64 are pattern matches that can be wrong on mixed or truncated content.
  4. Follow a suggested tool link to continue: a JSON payload leads to the formatter and schema tools, a timestamp to the Unix time converter, HTML to the escape and Markdown converters.
  5. Press Clear to reset the form. Editing the text after a detection also clears the previous result, so the card never shows a verdict for text that is no longer there.

What the detector checks, and what it does not

Formats it recognises

Unix timestamps in seconds (10 digits) and milliseconds (13 digits), ISO 8601 dates and date-times, JSON, JWT, cURL commands, URLs, HTML, XML, SQL, YAML, CSV, TSV, PEM blocks, UUIDs and Base64. Anything else falls back to plain text at 45%, which is the honest answer rather than a guess.

The rules run in a fixed order: timestamps and strict parsers first, then structural patterns such as HTML, YAML and delimited rows. A payload that satisfies more than one rule keeps the strongest match, so a JSON document containing commas is not reported as CSV.

What the confidence figure means

Confidence is the strength of the matched signal, not a probability. A JSON object that parses without error is reported at 99%, a payload that only looks like a comma-separated table sits at 87%, and unrecognised text sits at 45%. A bare number such as 12345678901 is valid JSON and is reported at 62%, because the tool cannot know whether it is an ID, a phone number or a timestamp.

Detection is not a security verdict. The page reads the text and never executes, fetches or decodes it, and it does not send the content anywhere, so a JWT, a private key or a customer sample can be inspected in place. Decoding and validation belong to the linked tools.

Large payloads, preview and limits

The character, line and byte counters follow the textarea, and the preview shows the first 2,200 characters. In testing, a 600 KB JSON document, a 1.7 MB CSV export and a 2.2 MB text file were classified in about a quarter to half a second each on a desktop browser, and the cost grows with the payload.

Only the format label, the evidence and a plain-text preview are produced here: Base64 is not decoded, JWT signatures are not verified, and XML or SQL are not checked for correctness. Those steps are the reason each result carries links to the tool that performs them.

Recent tools: