How does the link checker verify URLs?

The checker sends a HEAD request to each URL using the browser's Fetch API. A 2xx or 3xx response is reported as reachable. Network errors or 4xx/5xx responses are reported as broken or unreachable.

Why do some valid links show as "CORS blocked"?

Some servers do not include CORS headers, which browsers require for cross-origin fetch requests. A CORS block does not mean the link is broken — it means the server did not permit the check from a browser context. These links should be verified manually.

Are my links sent to a server?

No. The tool only sends requests directly from your browser to each URL you provide. No data passes through any jarvisbox server — the fetch is a direct browser-to-destination request.

What is the difference between "broken" and "unreachable"?

Broken means the server responded with a 4xx or 5xx HTTP status code (e.g. 404 Not Found). Unreachable means the request failed entirely — the host does not resolve, the connection timed out, or a network error occurred.

Does the checker follow redirects?

Yes. The Fetch API follows redirects by default (up to a browser-defined limit, typically 20). A redirect chain that ends in a 2xx response is reported as reachable.

Can I check internal (relative) links?

The checker works with absolute URLs only. Relative links like ../other-page.md cannot be resolved without knowing the base URL of your document.

Does the tool check image links too?

Only inline links ([text](url)) are checked. Image links (![alt](url)) are not included in the check. Use the Image Link Extractor to list image URLs separately.

Does this work offline?

Link extraction works offline. The checking step requires a network connection to reach each URL.

Markdown Link Checker

Paste a Markdown document, extract all links, and verify each URL is reachable before publishing.

100% client-side · fetches go directly to each URL

Markdown input

How to check links in Markdown

Paste your Markdown document into the text area.
Click Extract links to list all inline links found.
Click Check all links to verify each URL is reachable.
Links marked green are reachable. Red means broken or unreachable. Yellow means CORS-blocked (check manually).
Fix or remove any broken links before publishing your document.

Common use cases

README maintenance: Audit links in GitHub README files before a release.
Documentation audits: Find dead links across documentation pages.
Blog posts: Verify external links are still active before republishing older posts.
CHANGELOG files: Check that all linked issues, PRs, and commit hashes resolve.

常见问题

How does the link checker verify URLs?: The checker sends a HEAD request to each URL using the browser's Fetch API. A 2xx or 3xx response is reported as reachable. Network errors or 4xx/5xx responses are reported as broken or unreachable.
Why do some valid links show as "CORS blocked"?: Some servers do not include CORS headers, which browsers require for cross-origin fetch requests. A CORS block does not mean the link is broken — it means the server did not permit the check from a browser context. These links should be verified manually.
Are my links sent to a server?: No. The tool only sends requests directly from your browser to each URL you provide. No data passes through any jarvisbox server — the fetch is a direct browser-to-destination request.
What is the difference between "broken" and "unreachable"?: Broken means the server responded with a 4xx or 5xx HTTP status code (e.g. 404 Not Found). Unreachable means the request failed entirely — the host does not resolve, the connection timed out, or a network error occurred.
Does the checker follow redirects?: Yes. The Fetch API follows redirects by default (up to a browser-defined limit, typically 20). A redirect chain that ends in a 2xx response is reported as reachable.
Can I check internal (relative) links?: The checker works with absolute URLs only. Relative links like ../other-page.md cannot be resolved without knowing the base URL of your document.
Does the tool check image links too?: Only inline links ([text](url)) are checked. Image links (![alt](url)) are not included in the check. Use the Image Link Extractor to list image URLs separately.
Does this work offline?: Link extraction works offline. The checking step requires a network connection to reach each URL.

反馈这个工具的问题