How it works

Copy the HTML page, Markdown document, log line, email, or any other templated text you want to extract data from.

Replace the values you care about with EJS tags like <%= name %>.

Paste both below and get a clean JSON object with every value extracted.

Playground

Paste your rendered text and EJS template, then hit Extract.

Try an example:

Rendered text

EJS Template

Auto-run

Flexible whitespace match regardless of spacing

Strip whitespace remove leading/trailing whitespace per line

Safe mode return null instead of throwing on no match

Silent errors suppress template compilation warnings

Delimiter default: %

Open delimiter default: <

Close delimiter default: >

Partials (includes) reusable template fragments, as JSON: { "name": "<ejs source>" }

Type coercion convert extracted strings: { "age": "number", "active": "boolean" }

Extracted Data

Works with any text format - HTML, Markdown, logs, emails, CSV, plain text - anything you can describe with an EJS template
Loops & nested data - extract arrays and nested objects from any repeating section
Conditionals - if/else, switch/case, and boolean detection
Flexible whitespace - matches pretty-printed and minified HTML against the same template
Type coercion - convert extracted strings to numbers, booleans, or dates
Includes / partials - reuse template fragments across pages
Zero dependencies - 20KB, runs in any browser, Node.js, Bun, or Deno
Privacy-first - everything runs in your browser, your data never leaves your device