Reports, sharing and history
Where a finished AI Site Scanner report is stored, how to share its link, how a domain's earlier scans are listed, and what the public directory shows.
Every scan that loads the site is stored as a report with its own link, https://toolz.dev/ai-scan/r/<id>. The link shows the same report the scanner showed, plus the domain's earlier scans, and it keeps working after you close the tab. A scan that could not load the site is not stored and gets no link.
Where is my report stored?
On Toolz.dev. When a scan finishes, the scanner sends the result to the Toolz.dev backend, which stores it and returns a report id: 12 random, URL-safe characters, long enough that nobody can guess a link they were not given. The scanner page then shows Open this report on its own page, the command line prints Report: <url>, the MCP server returns Full report: <url>, and the HTTP endpoint returns the id as reportId.
The stored report keeps the domain, the URL scanned, the score, the grade, the site-type filter, the detected platform, whether it is public or private, when it was made, and the full scan result. A report never changes after it is written; scanning again makes a new one.
If storing fails (for example, the backend is briefly unreachable), the scan result is still returned and shown, and reportId is null. The scan is not held up waiting: the store step gives up after 4 seconds.
What does the report page show?
| Part | What it is |
|---|---|
| Header | The domain, the date and time of the scan in UTC, and a Copy link button (plus Share on a phone, which opens the system share sheet) |
| Report | The same grade, score, category cards, pages table and prioritized issues with Copy fix prompt buttons as the live scan |
| Earlier scans | The domain's public scans, newest first, up to ten, each with date, score and grade and linked to its report. Private scans are never listed here |
| Scan another site | A link back to the scanner |
Report pages are marked noindex, so search engines do not list them. They are meant to be sent to someone, not found. A link with an id that does not exist shows the site's not-found page.
How do I share a report?
Copy the link with Copy link or from the address bar, and send it. Anyone with the link can open the report, whether it was run as public or private; private only keeps it out of the public lists. Public and private scans, limits and Pro explains the difference.
How do I find a scan I ran earlier?
The scanner page has a scan activity block with two tabs:
- Recently scanned is public: the newest public scan of each domain, 10 on the scanner page, with grade, score, platform and date.
- Your scans is local: the last 8 reports run in this browser, one per domain, stored in the browser's local storage and never sent to the server. It includes private scans. Clearing site data clears it, and it does not follow you to another browser or device.
For older scans of the same domain, open any of its reports and use Earlier scans.
What is the public scan directory?
/ai-scan/scans lists the newest public scan of every site, newest first, up to 100. A domain scanned five times appears once, at its latest public result. Private scans never appear there, or in Recently scanned. Each row links to its report.
The Improved after a scan table on the scanner page is built from the same public scans: it compares each domain's first and latest public scan and lists those whose score rose. Fixing issues with fix prompts has the details.
What do the stats on the scanner page mean?
The row under the scanner shows, once there is at least one scan:
| Number | Meaning |
|---|---|
| Total scans | Every stored report, public and private |
| Average score | The mean score of every stored report, rounded |
| Not agent-ready | The share of stored reports that scored below 60 |
They count reports, not domains, so a site scanned ten times counts ten times. Scans that could not load the site are not stored and do not count.