Lumanu Readable text from any webpage.

Extract clean, AI-ready text from websites, raw files, and documentation. Built for developers, web scrapers, and LLM applications.

GET /extract
$ curl "https://lumanu.mcbcode.com/extract?url=https://example.com"
{
"success": true,
"text": "Example Domain. This page is for use in illustrative examples..."
}

What is Lumanu?

A simple edge-utility to convert bloated web markup into clean content blocks.

Clean Text & MD

Automatically filters out CSS sheets, layout scripts, headers, dynamic navigation menus, and yields clean developer-oriented plain text or markdown structures.

Perfect LLM Context

Saves tokens and reduces noise. Web pages are converted into clean markup blocks optimized directly for ingestion into vector databases and embedding pipelines.

Built on Workers

Powered by Cloudflare Workers at the edge. Highly responsive, minimal overhead routing, and extremely fast network connections close to target systems.

Interactive Sandbox

Test out the parser manually. Input a target URL to simulate real-time text extraction.

output.json
Ready. Input a URL and click "Extract Content" to run the parser.

Execution Boundaries

100
Max URLs per batch
5 MB
Maximum target size
30s
Connection timeout