Link Alpha is our featured choice for turning a broken-link report into a repair plan with ChatGPT, Claude, or Gemini. Check My Links is worth considering for routine page and document checks, Free Link Checker by LRT adds useful SEO link classifications, and Broken Link Checker from broken-links-checker.com offers a larger documented crawl allowance.
The best extension depends on the job. Reviewing references in one article, checking thousands of pages after a migration, and investigating a redirect chain require different capabilities. This guide compares five scanning extensions and one specialist companion, and explains how to interpret errors, avoid unnecessary changes, and turn findings into verified fixes.
Disclosure: Link Alpha and SEOdisaster share a developer. We feature Link Alpha's strengths while explaining where another tool may fit better. This is a comparison of published product documentation and store listings, not a controlled speed or accuracy benchmark. Features, limits, and pricing were researched on October 24, 2026 — check the linked product pages before installing or buying.
The best options at a glance
| Extension | Recommended use | Scope to understand | Pricing model at review |
|---|---|---|---|
| Link Alpha | Link checking followed by an AI repair brief | Current page; optional crawl of up to 25 same-origin pages | Free daily allowance; optional Pro |
| Check My Links (SelectorsHub) | Repeated page checks and Google Docs/Sheets links | Page or supported document | Free core; optional Advanced key |
| Broken Link Checker (broken-links-checker.com) | Larger browser-based crawls and document checking | Page, site crawl, Docs and Sheets | Three free checks, then paid |
| Free Link Checker by LRT | Inspecting internal/external links and link attributes | Page-level SEO link review | Free |
| Broken Link Checker (balvin.perrie) | Inspecting page links and links inside frames | Page/frame checking; documented deep-search mode | Installable without an upfront store charge; check current terms |
| Redirect Path (Ayima) | Diagnosing the URL you navigate to | Navigation and redirect inspection, rather than a bulk page scan | Installable without an upfront store charge; check current terms |
Recommendations are editorial judgments about documented capabilities. A longer feature list does not prove greater accuracy. Two extensions with the same name can also be entirely different products — use the publisher and installation links in this guide to identify them.
What should a broken link checker actually help you do?
A useful extension should help you discover a problem, understand it, locate its source, and verify the correction. Finding an error count is only the beginning.
Distinguish a destination from its occurrences
One missing URL can appear in a menu, a card, and an article paragraph, and across a site it may occur on many pages. You need to know both the failed destination and where it is used. Before choosing a tool, ask whether its report shows unique destinations, individual link occurrences, or both — otherwise “10 broken links” can mean very different amounts of repair work.
Separate redirects from errors
A working redirect is a different outcome from a missing destination, and it can be appropriate after a URL change. Check the final destination too: a redirect ending in an error is a different case from one ending on a relevant working page.
Show enough information to act
Useful report fields include the source page, target URL, anchor text, observed status, final destination when available, and a check timestamp. Export formats matter less than whether they preserve the information your editor or developer needs.
State what was not checked
A trustworthy workflow distinguishes an actual response from a timeout, inaccessible destination, excluded link type, or missing permission. Don't assume every red label means the resource has permanently disappeared.
1. Link Alpha: our featured option for AI-assisted repair planning
Link Alpha combines page scanning with locating links, local scan history, CSV/JSON exports, and an optional site scan. Its clearest distinction is the repair handoff: you can choose ChatGPT, Claude, or Gemini and open the assistant with a prepared brief populated for review.
This addresses a practical problem. A report might tell an editor that a link fails without explaining whether to replace the citation, correct an internal path, restore a page, or ask a developer to investigate access behavior.
Why the AI workflow is useful
The value is the context carried into the next step. A useful repair brief can keep the affected page and observed result together instead of asking you to reconstruct the problem in a new chat. Suppose an article references an outdated documentation path: the sensible decision depends on what the original sentence claims and whether the replacement documentation still supports that claim. AI can help organize that investigation and draft an edit, but the replacement still needs verification.
Link Alpha's documented workflow prepares the brief and can paste it into the selected assistant. You review it and submit it yourself — the extension does not automatically edit your website. Its privacy policy also describes a clipboard fallback if automatic insertion is unavailable. Treat “AI-assisted repair” as help with analysis and instructions, not proof that an AI-suggested destination exists or contains the right information.
What Link Alpha checks
The current product documentation describes checks of <a href> links, with on-page locating and status highlighting. It also describes checking ordinary same-page section links against the page's content. Hash-based application routes and text fragments require different treatment and may need review. It is not documented as an audit of every embedded image, script, or stylesheet — a linked PDF and an image displayed on the page are different kinds of item, so don't assume an extension checks both simply because it checks hyperlinks.
Site scanning: useful, but deliberately bounded
Link Alpha's site scan opens a dedicated Site Audit view and follows discovered links on the same origin. The documented limits are 25 pages per crawl and 10 site scans per day, shared by free and paid users. It follows robots.txt Disallow rules when choosing pages to read and does not use a sitemap for discovery, and a capped or incomplete crawl is labeled partial. That makes it a reasonable option for a small-site check or a focused sample; a site larger than the cap needs another approach for comprehensive coverage, and an upgrade does not remove those crawl limits.
Free and paid access
The product site lists 40 ordinary scans per day on the free tier, with exports and history included. Pro removes that daily scan cap and is listed at $3.99 monthly, $24.99 yearly, or $39 once. The site also documents a technical cap of 5,000 links per ordinary page scan. Compare daily scan limits separately from crawl page limits — they are different units.
Important limitation: read the reason behind the label
Link Alpha documents 401, 403, and 407 responses under Broken, while unresolved checks can appear as Could not check. A confirmed access refusal is not the same as confirmed deletion — open the details and reproduce the intended visitor journey before removing such a link.
Choose Link Alpha when you want page-level checking, usable exports, and a direct route into an AI repair conversation. Choose a larger crawler when the main requirement is exhaustive coverage beyond its published limits.
2. Check My Links: a strong option for routine page and document checks
The current SelectorsHub version of Check My Links highlights valid, redirected, and problematic links on the page. Its official materials also describe Google Docs and Sheets support, HTTP status reporting, domain exclusions, and CSV reporting. It is worth considering for an editor who wants to review a page's references without starting a broader technical audit.
The developer lists a free core version and an optional Advanced key priced at $29 once. The product page assigns features such as section exclusions, bulk rechecking, additional filtering, and Excel export to Advanced, though some features appear more broadly in the store description — check the actual free/paid boundary before buying. Older tutorials can also refer to different versions or installation IDs, so use the current SelectorsHub listing linked here.
Choose Check My Links when ongoing page or document checking matters more than a built-in handoff to a specific AI repair workflow. This comparison does not establish a documented whole-site crawl mode for this product.
3. Broken Link Checker from broken-links-checker.com: for larger browser-based crawls
This extension documents current-page checks, whole-site scanning, Google Docs/Sheets support, and CSV reports. Its site-scan documentation lists a default 1,000-page limit, configurable up to 10,000 pages per scan, and the store describes internal-link and sitemap discovery plus reporting of source pages and redirect information. Those published limits make it worth considering when Link Alpha's 25-page crawl cap is too restrictive — but they do not guarantee that every page on a large site will be discovered or checked successfully.
Its pricing page offers three free checks (a page scan or a whole-site scan counts as one check). Paid access is listed at $2.90 per month, $19 per year, or $29 once.
Choose this tool when you want a larger documented crawl allowance and source-page reporting while staying in Chrome. Evaluate discovery, completion, and actual output on your own site before relying on it for a migration.
4. Free Link Checker by LRT: for SEO link classifications
Free Link Checker by LinkResearchTools documents live, broken, and unverified link categories, internal/external distinctions, follow/nofollow inspection, source-code previews, and exports. It is a useful candidate when your questions extend beyond whether a URL responds — for example, whether a template marks links with an unexpected attribute, or whether a page points heavily to external resources.
The listing also advertises LRT's proprietary Power*Trust metric. Treat a vendor metric as supporting information, not a Google score or proof that a destination is trustworthy — a low proprietary score does not itself justify deleting a relevant citation.
Choose LRT when link attributes, scope, and source inspection are central to the task. The materials reviewed do not establish a comprehensive site-wide crawling workflow.
5. Broken Link Checker by balvin.perrie: for inspecting links in frames
This is a separate extension from the similarly named product above. Its store listing describes checking links in the top page and subframes, an inspect control that brings a link into view, status-code filters, keyword exclusions, and a same-domain deep-search option. It is a candidate for investigations where the location of links inside embedded frames matters, though frame handling is worth checking against the actual page, especially when an embedded system behaves differently from ordinary article content.
The listing reviewed showed a February 2023 update date. That doesn't prove the extension is broken or unsafe, but it is a reason to check current compatibility and support before standardizing on it.
Choose this tool when its documented frame inspection and filtering match your investigation. Validate the deep-search behavior and coverage rather than treating the description as an unlimited crawl guarantee.
6. Redirect Path: a specialist companion for destination diagnosis
Redirect Path from Ayima documents HTTP status and header inspection, including server redirects and client-side redirects such as meta and JavaScript redirects. It serves a different role from a scanner that checks all hyperlinks on a page: use it when you follow a suspicious link and need to understand the navigation path.
An illustrative investigation might start with an article URL redirecting to a broad homepage. The final response may succeed, yet the journey no longer delivers the promised reference. Inspecting the path helps establish what happened; editorial review establishes whether the destination is still useful.
Choose Redirect Path when you need to investigate an individual navigation or redirect behavior, and pair it with a scanning extension for discovering problematic links across a page.
Which extension fits your workflow?
For publishing and updating articles
Link Alpha and Check My Links are useful starting points. Prefer Link Alpha if a prepared AI repair brief would help you move from the report to an edit; Check My Links is worth considering if links in Docs or Sheets are part of the same editorial workflow. Regardless of the tool, verify that a replacement still supports the sentence around it — a working URL to unrelated material is not a successful citation repair.
For small website maintenance
Link Alpha's bounded site scan can help find recurring issues across a small group of pages. Check the completion status and page count: a partial scan is still useful evidence, but it is not proof that the remaining site has no errors.
For larger crawls and migrations
Compare the larger documented scope of broken-links-checker.com with a dedicated desktop crawler. Screaming Frog's official tutorial explains how to find failed URLs and their source inlinks, and its free version supports crawling up to 500 URLs — it is a desktop application, not a Chrome extension. Choose based on the larger job: a migration may also require indexability, canonical, metadata, and rendering checks, and link status alone does not establish migration readiness.
For link attributes and source inspection
Consider LRT. It is more relevant when you need classifications and HTML context than when the only question is whether a reference opens.
For redirect troubleshooting
Use a page scanner to discover candidates and Redirect Path to inspect individual journeys. Confirm the observed path, then decide whether the source link should point directly to the relevant destination.
How to read a link report accurately
HTTP statuses describe the response to a particular request. They need interpretation in the context of the page and intended audience.
| Result | What it usually means | Sensible next step |
|---|---|---|
| 200 | Request succeeded | Confirm the content still matches the link's promise |
| 301 or 308 | Permanent redirect | Check the final destination and relevance |
| 302 or 307 | Temporary redirect | Decide whether the behavior is expected |
| 401 | Authentication is required | Test the intended access journey |
| 403 | Access was refused | Investigate permissions or request blocking |
| 404 | Resource was not found | Verify the address; find a relevant replacement if needed |
| 410 | Resource is reported as gone | Remove or replace the reference where appropriate |
| 429 | Too many requests | Reduce activity and retry later |
| 5xx | Server-side failure | Recheck and investigate whether it persists |
| Timeout or unreadable response | No conclusive response was obtained | Investigate the reason before declaring deletion |
A 200 response is not proof of a useful page — it might show an error message or irrelevant content, and Google also distinguishes soft 404 situations and says successful responses do not guarantee indexing. Likewise, a browser extension's request is not a Googlebot test: different clients, access states, and request methods can produce different outcomes.
Why a link can open manually but fail a scan
Possible explanations include authentication, request blocking, rate limits, timing, or browser permissions. Cross-origin response access is also governed by browser security rules, and an extension's execution context and permissions matter — though CORS does not prevent every link checker from checking external links.
Record the actual evidence. “403 returned to the check” is more precise than “page deleted.” Don't bypass a site's access controls to force a green result.
Why two tools can report different totals
They may count unique URLs or repeated occurrences, scan different frames, exclude non-HTTP links, use different discovery methods, or run before all relevant links appear. A larger total does not automatically mean better detection — compare the actual URLs and page locations, check what each tool skipped, and note whether the page changed between scans.
A reliable workflow: find, review, repair, and recheck
Step 1: choose the page state
Open the version of the page you want to audit. Let relevant content load and expand navigation or sections if needed. Record whether you are signed in and whether the page is public, staging, or personalized — a result from an authenticated preview doesn't necessarily represent the experience of a public visitor.
Step 2: scan and save the evidence
Run the extension and let the check finish. Record the source URL, time, completion status, exclusions, and any crawl cap, and export if you need to hand the work to someone else.
Step 3: prioritize useful repairs
Start with a failed primary action, navigation link, or important reference. A repeated template issue may affect more visitors than a one-off footnote, so don't prioritize solely by raw error count — ten occurrences of the same footer link can often be fixed in one template change.
Step 4: verify before changing
Open the destination, inspect its response where needed, and compare what it delivers with the surrounding text. Recheck uncertain or temporary results.
Step 5: choose a repair that preserves meaning
| Situation | Appropriate investigation or repair |
|---|---|
| Typo in an internal path | Correct the source link |
| Page moved to an equivalent destination | Update internal links; maintain an appropriate redirect |
| Reference withdrawn without an equivalent | Find another source or revise the unsupported statement |
| Broken link in shared navigation | Correct the shared component and check affected templates |
| Access-controlled destination | Make the access requirement clear or choose a suitable public reference |
| Temporary service failure | Recheck before replacing a useful source |
Google supports permanent redirects for permanent moves. Avoid redirecting every missing resource to the homepage — the destination should fit the original purpose.
Step 6: recheck the edited source
Validate the deployed page, not only the replacement destination — the CMS may still contain an old link in a button, menu, or duplicate occurrence. Record the correction and outcome, and keep unresolved cases assigned to an owner instead of hiding them in an export.
Using Link Alpha's AI brief effectively
The best AI request asks for a decision supported by evidence, rather than a list of guessed replacement URLs. After opening a prepared repair brief, add context that only you know: whether you control the destination, whether the page moved, what the citation was intended to support, and what changes you can make in the CMS.
For a citation problem, you could add:
“This link supports the sentence immediately before it. Find a current primary source that supports the same claim. Open and verify any proposed destination. If the evidence no longer supports the claim, recommend a revision instead of inventing a replacement URL.”
For an internal routing issue:
“I control this website. Distinguish a typo in the source from a missing redirect or an intentionally removed page. Recommend the smallest appropriate correction and tell me how to verify both the source link and final destination.”
Those are suggested additions, not claims about the exact text generated by Link Alpha. Review sensitive information before using an external assistant: the policy describes local scan storage and a user-triggered transfer into the chosen AI interface, and once you submit information to an AI service, that service's terms apply. Automatic redaction should not replace your own review.
Broken link checking and SEO: keep the claims realistic
Google uses links to discover pages and understand relevance, and recommends crawlable links with descriptive anchor text. Repairing important internal navigation and maintaining useful references supports a functioning site. It does not follow that every external 404 causes a ranking penalty or that installing a checker improves rankings — an extension reports observations; it does not award authority or prove that Google has indexed the destination.
Assess repaired user journeys alongside technical outcomes. For broader indexing questions, use the guides below.
What no extension can guarantee
A link scan cannot guarantee complete coverage of undiscovered pages, continued availability tomorrow, relevance of every successful response, or successful completion of a purchase or form submission. Fragment validation also matters: a page can respond successfully while its linked section is missing, so only count fragment support as confirmed when the tool explicitly validates that behavior, and test complex application routes separately.
For critical calls to action, follow the actual journey. For recurring monitoring, confirm that the chosen product really supports scheduling and alerts — a saved manual scan is not scheduled monitoring.
Permissions and privacy deserve a quick review
Check the publisher, requested access, current privacy policy, and whether a feature sends data to an external service. Chrome lets you manage extension site access, but changing it can affect which checks are possible. A store privacy declaration is a developer statement, not an independent security audit, and “runs locally” does not mean no network requests — checking a destination generally requires contacting that destination.
A practical checklist for comparing extensions yourself
Use a small test page you control with known examples: a working destination, a deliberate 404, a working redirect, a redirect ending in an error, a repeated URL, and a missing section anchor. If relevant, add links inside frames or dynamically expanded content.
| Question | What to record |
|---|---|
| Did it find the known links? | Expected and observed URLs, not just totals |
| Did it explain uncertain checks? | Status, reason, and retry behavior |
| Did it locate the source? | Whether the correct occurrence was highlighted |
| Did it preserve useful context? | Export columns and source-page information |
| Was the scope clear? | Checked pages, skipped pages, limits, completion state |
| Did it help complete the repair? | Steps needed to edit and verify the source |
This is a suggested evaluation protocol — we did not perform or invent results for it in this article. It is more meaningful than choosing a tool from star ratings or an unsupported claim of perfect accuracy.
Our recommendation
Start with Link Alpha if you want to connect finding a problem with planning its repair. Its exports, locating workflow, and prepared ChatGPT, Claude, or Gemini brief make it a useful option for editorial maintenance and smaller checks.
Choose another tool when its documented scope fits better: Check My Links for supported document workflows, LRT for detailed link classifications, a larger scanner for wider crawl coverage, and Redirect Path for navigation diagnosis. Whichever you choose, measure success by a corrected and verified user journey, not by a green counter alone.