All workflows
RDAP logoWayback Machine logoAnthropic logoSerper logoGoogle Sheets logo

Look up an unfamiliar website before you deal with it

Gathers what public records, the web archive and search results say about a site you don't know, each with its source.

Run manually5 apps5 steps

Apps it connects

Authenticate each one once and the workflow is ready to run.

What it does

Each integration step this workflow runs, in order, what it is for, and the conditions and loops that decide when it runs.

  1. 1RDAPLookup domainLooks up when the domain was registered, through which registrar, and the registrant as the public registry publishes it.
  2. 2Wayback MachineGet capture historyChecks when the Internet Archive first captured the address and how many captures it holds in total.
  3. If pageReached is true and pageText is not empty

    1. 3AnthropicGenerate model responseAsks the AI to quote the address, phone, email and company details the site's own page states, and nothing more.
  4. 4SerperSearch webSearches for the domain to find results on other domains, then searches again for the site's reviews.
  5. For each reportRow

    1. 5Google SheetsAppend rowAppends one row per signal with the date, site, what was found and the source it came from.

How it works

What happens on each run, in plain English.

Give it the address of a website you do not know, and it gathers what is publicly on record about it before you buy from it, apply to it or hand over your details. It checks the domain's registration record, its history in the Internet Archive, the contact details stated on its own page, which top search results sit on other domains, and what a search for its reviews returns.

You end up with five rows appended to your spreadsheet, one per signal, each naming its source so you can open it and check for yourself. It reports and does not conclude: where nothing was found, the row says so rather than treating an absence as a finding. The AI is asked only to quote the site's page, and if its answer strays into judgement the row says so instead. Only what any visitor can see is read, and the site itself gets a single page request and nothing more.

Before the first run you need a tab in your spreadsheet headed 'Checked on', 'Site', 'Signal', 'What was found' and 'Source', in that order.

websitedomaindue diligenceresearchvendorgoogle sheetsai

Start from a workflow that already works.

Add "Look up an unfamiliar website before you deal with it" to your workspace, connect its apps, and make it yours.

Free plan, no credit card