| 开发者 | dragoncoreltd |
|---|---|
| 更新时间 | 2026年10月10日 12:05 |
| PHP版本: | 8.0 及以上 |
| WordPress版本: | 7.1 |
| 版权: | GPL-2.0-or-later |
| 版权网址: | 版权信息 |
wp dragon-broken-links scan, list and recheck for large or headless sites.dragon-broken-links folder to /wp-content/plugins/.No. There is no cloud account, no API key, no per-link credits, and nothing is reported to Dragon Core or anyone else. The only outbound traffic is the request your server makes to each external link to see whether it still works (see "External services" above).
No. Every unique URL is checked once per cycle, however many posts link to it, and there is no cap on links, posts or scans.
Broken - failed on three separate checks with a hard error (404, 410, no such domain, connection refused). Suspected - failed once; it is re-checked before it counts as broken. Needs review - the site answered but ambiguously (403, 429, timeout, server error), which is often bot-blocking rather than a dead page; open it, and dismiss it if it works.
Some servers answer automated requests with a 403, 429 or a timeout even though the page is fine for real visitors. Rather than cry wolf, Dragon Broken Links flags those for you to decide, and never counts them as broken automatically.
No. External links are checked in small, throttled batches within a fixed time budget, and each unique URL is checked only once. Internal links are resolved from the database rather than by making requests to your own site.
Yes - <img> sources, plus iframes, video, audio and embeds in your posts and pages. Internal images are checked against the uploads folder (including resized variants); external ones are requested like any other link. Gallery thumbnails and srcset variants of an image that exists are not checked separately.
Yes. Links are re-checked automatically in the background (every 15 minutes, working through every link daily, weekly or monthly - your choice under Settings), and a post is re-scanned whenever you save it. On top of that, every post is read again on a schedule - weekly by default, or daily, or off - under Settings → Full re-scan.
Yes. Every public post type is scanned by default - posts, pages, WooCommerce products and any custom post type your theme or plugins register. Untick the ones you do not want under Settings → Content scanned.
Yes, if you switch it on: tick Custom fields and/or Approved comments under Settings → Also scan. Links found there are reported with the post they belong to. Edit URL, Remove link and domain replace only change post content, so fix those links where they are stored (the report tells you when a link is in a custom field or comment).
Yes. Tick Email alerts under Settings. When a scan finds links that have newly become broken, you get one email listing them, sent to the site admin email or to the addresses you enter. Links that were already broken are not sent again, and no email goes out when nothing new broke. A link that breaks when you re-check it by hand is sent straight away, or with the running scan's email if a scan is in progress. The email uses your site's normal mail setup, so if other WordPress emails reach you, these will too.
Yes: Tools → Broken Links → Bulk tools → Replace a domain points every broken, suspected and needs-review link on the old domain at the new one. Working links are left alone. Links it cannot change are counted in the result, and pages built with a page builder are named.
Yes: Bulk tools → Export findings (CSV) downloads the broken and needs-review links, one row for each place a link is used.
Yes. If Broken Link Checker's data is still in your database, Settings → Import from Broken Link Checker carries your dismissed links over so you do not have to re-triage them. Nothing is removed from Broken Link Checker.
Background scanning uses WP-Cron, which only runs when your site gets traffic. On a quiet site, click Scan now, or run wp dragon-broken-links scan from WP-CLI. If WP-Cron is turned off on your site, the report says so: background checks then run only when your server calls wp-cron.php on a schedule.
Yes. Fixes are applied through the block parser, updating both the block data and the rendered HTML together, so the editor never reports invalid content. Classic (non-block) content is handled too, and on posts and pages every fix creates a revision you can restore from (some post types, such as WooCommerce products, keep no revisions). Removing a link from a button keeps the button and its text. Pages built with Elementor, Beaver Builder or Bricks keep their own copy of the content, so their links are fixed in the builder; the report says so and offers to open the page.
Yes, per site: each site keeps its own report, settings and background re-check.
&, -- or backslashes lost their backslash on save). Every fix is now saved exactly as the editor would.