Bulk SEO & Broken Link Checker — DigestSEO

Audit a list of public webpages for technical SEO issues, markup-based accessibility signals, JSON-LD errors, and selected broken-link observations. Get a clear page-by-page dataset plus a downloadable issue report. No external SEO API key is required.

Best for: bulk pre-launch checks, website migration QA, content audits, and checking a specific list of URLs. Submit 1–1,000 public HTTP/HTTPS page URLs per run. You pay for completed page audits under the current Apify Store event pricing; incomplete/skipped/failed pages do not trigger the page-audit event.

Quick start

1. Open the Actor on Apify and go to Input.
2. Add public webpage URLs in the Webpage URLs field (or use JSON).
3. Keep Check links enabled to inspect selected links. Choose Maximum links per page between 1 and 20 (default: 10).
4. Run the Actor. For large inputs, configure enough time in Run options (for example, 3,600 seconds for a thousand-page test) and choose a spending limit you're comfortable with.
5. Open Output for page results and issue exports. Inspect Run summary and Continuation to see what was actually processed.

Example input

{
  "urls": [
    "https://digestseo.com/",
    "https://digestseo.com/keyword-research/"
  ],
  "checkLinks": true,
  "maxLinksPerPage": 10
}

The Actor accepts an explicit URL list. It does not discover a sitemap, crawl an entire domain, or follow links to add new pages. Link checks inspect up to 20 selected candidates per page, giving main content links priority; they are not exhaustive.

What gets checked

• Technical SEO metadata: titles, meta descriptions, canonicals, index directives, headings, and other checks supported by the pinned validator.
• Basic accessibility signals: markup-level observations such as missing accessible image names. Not a complete WCAG audit.
• JSON-LD markup: syntax and document-shape issues. Not a full Schema.org or rich-result eligibility test.
• Selected link responses: reachable, confirmed not-found, redirected, restricted, rate-limited, transient, unreachable, and skipped-by-policy. A HEAD 404/410 is confirmed using bounded GET before being treated as broken.

Download your results

Output	What it contains
Page audit results (default dataset)	One result per unique submitted URL, including status, findings, link observations, and coverage flags.
Actionable issues (CSV) — ISSUES.csv	Spreadsheet-friendly rows with affected page, issue severity/code, linked target when applicable, observed HTTP status, and an action to review.
Actionable issues (JSON) — ISSUES.json	The same prioritized issues as structured data, plus omission counts and coverage/truncation disclosures.
Run summary — SUMMARY.json	Completed, failed, skipped, partial and unprocessed page counts; limits, stop reason and charged-event count.
Continue unfinished audit — CONTINUATION.json	Always states whether work remains. When incomplete it contains an input object with only unprocessed URLs.
Next batch input — NEXT_BATCH_INPUT.json	Available in Storage → Key-value store only when URLs remain; copy its entire JSON into a new run's input.

The issue export includes reported page/markup findings, relevant link observations and execution errors. Access restrictions, redirects, timeouts and robots-skipped links are not automatically broken links. Recommendations are review prompts, not changes made to your website. Error-level issues appear first in the export. If result sizes are capped, ISSUES.json explicitly reports omitted rows and affected pages; consult the original dataset for detail.

When your run stops early

A large input may stop at the run's time or event-spending limit without failing. Open Storage → Key-value store → NEXT_BATCH_INPUT.json and paste that JSON into a new run. It preserves checkLinks and maxLinksPerPage and contains only URLs that never received a persisted page result. Already processed pages, including partial or failed outcomes, are not repeated automatically. Review SUMMARY.json to understand why processing stopped.

If you resurrect the same run instead, its checkpoint protects already-persisted results and charge replay. Never remove or edit that run's checkpoint manually. A completed run removes any stale NEXT_BATCH_INPUT.json.

Interpret findings correctly

• Not found: bounded checks returned 404/410; verify before updating your source link.
• Redirect: the linked URL redirects; its final destination was not followed or validated.
• Restricted / rate-limited / unreachable / transient: access or availability couldn't be confirmed. Do not automatically delete those links.
• Skipped: the target was excluded by network/robots policy; no HTTP link probe was made.
• Metadata length notices: editorial hints, not ranking penalties or a requirement to extend text to a fixed length.

Important limits and privacy

Public webpages only. No login, cookies, private-network access, browser/JavaScript rendering, screenshot capture, authenticated crawling, proxy bypass, AI-generated fixes, SERP/keyword/rank/backlink data, or SEO ranking guarantees. Remote robots.txt and public-network restrictions are respected.

The Actor does not store full fetched HTML documents, credentials, cookies, or screenshots and does not send results to DigestSEO. Individual findings may contain short evidence snippets from page markup. Run inputs and results are stored by Apify according to your account's platform retention and data settings.

For support and additional tools visit DigestSEO or DigestSEO support.

Developer and security implementation details are maintained separately from the customer-facing Store README.