SEO issues that arrive as pull requests, not spreadsheets
Connect the repo, crawl the site, review the diff. CrawlX finds the template behind thousands of errors and opens the fix on GitHub — you stay in control of what merges.
Most SEO tooling is designed to be read by marketers and forwarded to engineers. CrawlX is designed to be actioned. It crawls the deployed site in the cloud, runs 150+ technical checks with the evidence attached to each finding, collapses thousands of page-level errors into the one template that causes them, and — when the repository is connected — drafts the change as a pull request against a branch. There is a REST API to trigger crawls and pull results as JSON, signed webhooks when a crawl finishes, and JavaScript rendering so client-side apps are audited as Googlebot would see them.
What hurts today, and what CrawlX does about it
The SEO team sends a spreadsheet with 3,000 URLs and the same error on each.
Root-cause grouping folds page-level findings into the template or component that produces them. You see one issue affecting 3,000 pages, with sample URLs and the exact tag or header that triggered it.
You are told what is wrong, never what to change.
Connect a GitHub repository and CrawlX drafts the fix as a pull request — a readable diff, on its own branch, never pushed to your default branch. You review, edit and merge like any other PR. AI drafting uses your own Anthropic, OpenAI or Gemini key.
Regressions ship silently and surface weeks later in Search Console.
Scheduled crawls recrawl on a cadence, crawl comparison diffs the runs, and signed webhooks fire when a crawl completes so you can route the payload to whatever you already use for alerts.
Half the page only exists after hydration.
JavaScript rendering (Professional and above) crawls the rendered DOM in headless Chromium and flags content, links and metadata that are missing from the initial HTML response.
Every tool wants you to live in its dashboard.
The REST API on Professional and above starts crawls and returns issues as JSON, so results can land in a CI job, a ticketing system or a script — with API keys scoped per project.
Redirect chains, canonical loops and hreflang mismatches that nobody owns.
150+ technical checks run on every crawl, each with the response chain, header or tag that produced the finding, so a ticket carries proof rather than a hunch.
Three steps, one loop
Connect the repository
Install the CrawlX GitHub App on the repo that builds the site. CrawlX only ever opens pull requests — it cannot push to your default branch.
Crawl the deployed site
Point CrawlX at production or a preview URL. Turn on JavaScript rendering for client-side apps. Results stream in real time and group by root cause.
Review the pull request
For high-impact issues, CrawlX drafts the change and opens a PR with the affected page count in the description. Merge it, or use it as a starting point.
Pick the plan for developers
Every plan runs the same technical checks — the difference is crawl size, integrations and seats.
Free
Free500 URLs per crawl, 5 crawls a month and 1 project — enough to audit a side project or a docs site and see how findings are grouped.
- 500 URLs per crawl
- 5 crawls / month
- 1 project
- 150+ SEO checks
- CSV / JSON / PDF export
Starter
$29/moScheduled crawls, Search Console + GA4, and bring-your-own-key AI fix suggestions so findings turn into drafted changes.
- 10,000 URLs per crawl
- White-label reports
- Search Console + GA4
- Scheduled crawls
- AI fix suggestions (BYOK)
Professional
$79/moJavaScript rendering for client-side apps, the REST API and signed webhooks for CI, custom extraction, and up to 5 seats for the team.
- 100,000 URLs per crawl
- JavaScript rendering
- REST API + webhooks
- Custom extraction
- Up to 5 team members
Annual billing is cheaper — see the full breakdown on the pricing page.
Common questions
Does CrawlX push commits to my repository?
No. CrawlX opens pull requests on a separate branch through the CrawlX GitHub App and never pushes to your default branch. Nothing merges without a human approving it.
Can I run CrawlX from CI?
Yes, on Professional and above. The REST API starts crawls and returns findings as JSON, and signed webhooks notify your endpoint when a crawl completes, so a pipeline can trigger a crawl on deploy and react to the result.
Does CrawlX render JavaScript?
Yes. On Professional and above CrawlX renders pages in headless Chromium and crawls the rendered DOM, flagging content and links that only exist after hydration. Plans below that crawl the raw HTML response.
Which AI model drafts the fixes?
Whichever one you bring. AI features are bring-your-own-key: plug in an Anthropic, OpenAI or Google Gemini key and CrawlX uses it for fix drafting, so you control cost and there is no vendor lock-in.
CrawlX for other teams
See it on your own site
Your first cloud crawl runs free — 500 URLs, no credit card, no install.