# Crawlwise > Audit your site for SEO and AI search. Crawlwise crawls up to 25 pages, checks what search engines and AI assistants need to find and understand them, and tells you which fixes matter most. ## What it is Crawlwise is a free AI SEO audit tool. Give it a domain and it crawls up to 25 publicly reachable pages, reads 25 of them with a language model, and returns a ranked action plan: what to change, on which URL, in what order, with the example copy to change it to. It also reports whether an answer engine could quote each page. No account, no card, no plugin or script to install. ## Facts - Price: free. 3 full AI audits per day; technical audits are unlimited. - Pages crawled per audit: 25. Pages read by the model: 25. - Time for a full audit: about two minutes. - Crawl rate: at least 1 second between requests to the same host; robots.txt and Crawl-delay obeyed. - Account required: no. Card required: no. - Not built yet: competitor comparison, scheduled re-scans, keyword positions, regression alerts, JavaScript rendering. ## What it measures - Titles, headings, meta descriptions, structured data types, content depth. - Whether AI crawlers are blocked in robots.txt, and whether /llms.txt exists. - Citability: whether a page opens with a self-contained definition of its subject, whether its answers survive being quoted out of context, and whether it offers lists or tables an answer engine can lift. ## Questions ### What is Crawlwise? Crawlwise is an AI SEO audit tool. It crawls up to 25 pages of a public website, reads each one with a language model, and returns a ranked list of what to change — with the example copy to change it to. An audit takes about two minutes and needs no account. ### Is Crawlwise free? Do I need an account? Crawlwise is free and needs no account and no card. Technical audits are unlimited, and 3 full AI audits run per day. ### Do I need to install anything on my site? No. We crawl publicly reachable pages the same way a search engine does — no plugin, no script, no account. Pages behind a login are not reachable and are skipped. ### How is this different from a classic SEO crawler? A crawler compares your tags to generic rules. Crawlwise reads the actual copy, works out what each page is for, and writes advice specific to your product — including the sentences to use. ### Will this slow my site down? No. We read robots.txt before the first page, honour Crawl-delay, and space requests one second apart per host — concurrent audits of the same site share that pace rather than doubling it. A 25-page crawl is about 30 seconds of light traffic. ### What does the audit not cover yet? Competitor comparison, scheduled re-scans, keyword positions and regression alerts are not built. Neither is JavaScript rendering — a page that ships its copy only from the client will read as thin. ### Does the AI make things up? Every recommendation is anchored to a specific URL and the crawled evidence behind it — the page count, the measured value, the exact tag. If we can’t measure it, we don’t claim it. ### How long does an audit take? A Crawlwise audit takes about two minutes for a 25-page site. The crawl itself is roughly 30 seconds, because requests are spaced 1 second apart per host; the rest is the language model reading the pages. ### What does Crawlwise check? Crawlwise measures titles, headings, meta descriptions, structured data types and content depth on every page it reads. Each is judged against what the page is actually for rather than against a generic rule — a thin title on a pricing page and a thin title on an article are not the same problem. ### Can AI assistants like ChatGPT and Claude cite my site? Crawlwise reports whether an answer engine could quote each page: whether the page opens with a self-contained definition of its subject, whether its answers survive being quoted out of context, and whether it offers lists or tables an engine can lift. It also checks robots.txt for GPTBot, ClaudeBot, PerplexityBot, Google-Extended and OAI-SearchBot, and flags a missing /llms.txt. ### Can I hand the report to a coding agent? Every Crawlwise report includes a prompt written for a coding agent. It carries the findings, the URLs and the evidence behind each one, so an agent can act on the report directly instead of you retyping it. ### What data does Crawlwise store? Crawlwise stores the measurements taken from your pages, not the text of them. It does not collect email addresses, and does not resell what it crawls. ### How do I stop Crawlwise crawling my site? Two lines in robots.txt stop CrawlwiseBot: a User-agent line naming CrawlwiseBot, and Disallow followed by a slash. CrawlwiseBot reads robots.txt before requesting any page and obeys it — no account and no request to us needed. ## Pages - [Crawlwise](https://crawlwise.app/): the audit form, how it works and FAQ. - [CrawlwiseBot](https://crawlwise.app/bot): what our crawler is, how it behaves, how to slow it down or block it. ## Our own crawler CrawlwiseBot identifies itself as `CrawlwiseBot/0.1 (+https://crawlwise.app/bot)`. It reads robots.txt before requesting any page and obeys it. Two lines in robots.txt stop it: a User-agent line naming CrawlwiseBot, and Disallow followed by a slash. ## Crawlers we allow - GPTBot - OAI-SearchBot - ClaudeBot - Claude-User - PerplexityBot - Google-Extended - CCBot Last updated: 2026-09-29