Compress HTML
AI / AgentStrip scripts, styles, comments, classes, and whitespace from HTML to minimize token usage before LLM ingestion.
A scraped page's HTML markup is mostly noise from an LLM's perspective (scripts, style blocks, class and id attributes that add no semantic value), and this strips all of that while preserving visible text content and structural tags, which typically cuts token usage significantly before you pass the page into a prompt or vector database. Reduction varies a lot by page: heavily styled pages with inline scripts can shrink 60-90%, while already-minimal HTML barely changes. It's a different goal from the HTML Minifier tool, which optimizes for byte size while keeping the page visually renderable. This output isn't meant to render correctly in a browser, since scripts, styles, and classes needed for visual fidelity are gone. Comparing before/after character counts is a quick way to see how much of your token budget was going to markup rather than content.
How to use Compress HTML
- 1.Paste a full HTML page to strip out scripts, styles, comments, and class attributes before feeding it to an LLM as context, cutting token usage significantly.
- 2.Use it to shrink scraped page HTML before storing it in a vector database or passing it to a summarization pipeline.
- 3.Compare the before/after character count to see how much of your token budget was being wasted on markup noise.
Frequently asked questions
Use via API, SDK, or MCP
cURL# Free: 1,000 req/day · Pro: 10,000 req/day
curl -X POST https://api.utilix.tech/v1/tools/compress-html \
-H "Authorization: Bearer utx_live_..." \
-H "Content-Type: application/json" \
-d '{"html":"<html>\n <!-- nav -->\n <body>\n <h1>Hello</h1>\n </body>\n</html>","removeComments":true,"removeScripts":true,"removeStyles":true,"collapseWhitespace":true}'Get an API key from your dashboard · Full API docs →