Free tools
Free web data tools, no sign-up.
Paste a public page and take what you need from it, in your browser. Five a day, and when a page can’t be read, you get the reason.
The tools
- URL to MarkdownPaste a public page and get clean Markdown for an AI agent, a RAG pipeline or your notes.
- Link ExtractorEvery link on a page, up to 500, ready to copy or download.
- Meta Tag CheckerTitle, description, language and canonical URL, as the page states them.
- URL to JSONName up to 20 fields and get them as JSON, each with where on the page it was read.
- PDF to MarkdownA PDF’s text, page by page, as Markdown, for files up to 5 MiB.
HOW THEY WORK
Paste, read, take.
Paste a public URL
Any page anyone can open without signing in. No account, no key, no install.
Octocrawl checks, then reads
It reads the site’s robots.txt first, then the page over HTTP, and keeps what the tool is for.
Copy or download
Take the result as text or a file. If the page can’t be read, the reason comes instead of an empty file.
FROM YOUR CODE OR AGENT
The same reads, one call away.
For more than five a day, a whole site, or inside your agent.
REST API
Hosted, no key for 20 pages a day; a free account’s key for 1,000 credits a month.
curl https://api.octocrawl.dev/v1/scrape -H 'content-type: application/json' -d '{"url":"https://example.com","formats":["markdown"]}'API referenceMCP
Claude Code, Cursor, Codex and OpenCode read pages through Octocrawl’s tools.
claude mcp add --transport http octocrawl https://mcp.octocrawl.dev/mcpConnect MCPCLI
On your own computer, with no daily limit, plus crawl and batch.
npx octocrawl scrape https://example.com --markdownAdvanced referencePlayground
Signed in, run any page on your credits, read in a browser when it needs one.
octocrawl.dev/dashboard/scrape/Open the Playground
FAQ
Free tools,
answered plainly.
Are these tools free?
Yes. Every tool here runs in your browser without an account: five pages a day per visitor (at most 150 a day for the whole site), reset at 00:00 UTC.
Do I need an account?
No. Sign in when you want more: a free account gets 1,000 credits a month for the hosted API, MCP and the Playground, which also reads pages that need a browser. No card.
Can I use them from code or an agent?
Yes. The same reads are one call to the hosted API (api.octocrawl.dev) or the MCP server (mcp.octocrawl.dev), or run npx octocrawl serve on your own computer with no limit.
What happens when a page can’t be read?
You get the reason instead of an empty file: the site’s robots.txt disallows it, it blocks the request, it needs a login, or it draws its content with JavaScript, which these free tools don’t run.