# URL extractor: run it free, or from your agent

Every URL in pasted HTML or text: links, images, scripts and emails, with internal and external filters and CSV.

## Prompt

Use looot to list every outbound link on these pages: [page URLs]. 1. Set up https://looot.ai/skill.md. 2. Find a web page to markdown endpoint in the catalog and check its price. 3. Fetch each page once. 4. Pull every link from the markdown and keep the ones on another domain. 5. Return a table of page, link and anchor text, and the total cost.

## What it needs

Page HTML, Markdown or text, and an optional page URL

## What comes back

- URL
- type
- anchor text
- internal or external
- count

## Endpoints and providers

- context-dev: context-dev-web-scrape-markdown

## What it costs

Nothing. The tool on the web page runs in the visitor's browser: no account, no API call and no limit per hour. The prompt above runs the same job from your own looot account with current data, billed per call.

## Audit internal links

See every page a URL links to, and which links leave the site.

## Check outbound links

List the external domains a page sends visitors to before you publish it.

## Collect asset URLs

Pull every image, script and stylesheet URL out of a template.

## Clean up a paste

Turn an email, a doc or a chat log full of links into one deduped list.

## How do I extract all URLs from a web page?

Open the page, view its source with Ctrl+U or Cmd+Option+U, select all and paste it in the box at the top of this page. Every URL appears in the table, deduped, with a count.

## Can it fetch a page from its URL?

No. This page reads only what you paste, so it never loads another site. The page URL field is used to turn relative links into full ones. To fetch many pages, give the prompt above to your agent.

## Which URLs does it find?

href, src, srcset, action and poster attributes, link and meta tags such as og:image, mailto and tel links, and bare http, https and www. links in the text. javascript: and data: values and links to a spot on the same page (#top) are skipped.

## What counts as internal?

A relative link such as /pricing, or a full link on the same host as the page URL you typed. www. is ignored, so www.example.com and example.com match. A subdomain such as docs.example.com counts as external.

## Does it work on Markdown and plain text?

Yes. Links in Markdown and in any text are found as bare URLs. Relative paths need HTML tags around them to be found.

## How do I export the list?

Copy list puts one URL per line on your clipboard. CSV adds the type, anchor text, internal flag and count. TXT saves the plain list.

## Is there a size limit?

It reads the first 2 million characters and keeps up to 5,000 distinct URLs. The table draws the first 300, and copy and download hold them all.

## Is it free and private?

Yes. It runs in your browser with no account and no limit, and what you paste never leaves your device.

## Related tools

- /free-tools/website-to-markdown
- /free-tools/email-extractor
- /free-tools/sitemap-xml-generator
- /free-tools/utm-builder

## Last reviewed

2026-10-16.

## Agent skill for this page

https://looot.ai/free-tools/url-extractor/skill.md sets an agent up for exactly the jobs on this page, with the live price of each, and offers the page's starter task once the user has picked a job and seen the price. The line to give an agent is under "Give this to your agent" below.

## For agents

### Give this to your agent

```text
set up looot for URL extractor - https://looot.ai/free-tools/url-extractor/skill.md
```