Skip to content

Scrapier returns publicly accessible data at your instruction. You are responsible for what you ask it to collect and what you do with the results. You may use it to research products, prices, places, content and public profiles, and to build your own tools on top of that data.

You may not use it to:

  • stalk, harass, surveil or track a person, or find someone who doesn’t want to be found
  • compile profiles of individuals in order to sell them, or to target or discriminate against people
  • identify people from anonymous activity
  • collect data that is private, behind a login you aren’t entitled to use, or that you have no lawful basis to process
  • break the law, or the terms of the site you scrape, where that is your responsibility
  • send abusive volumes of requests, or evade limits to harm a site

The same rules apply when an AI assistant calls Scrapier for you through the Claude connector, ChatGPT & Codex or the MCP server. We may suspend access that breaks them.

If you want data about you or your site excluded, or you think Scrapier is being misused, email [email protected] with the page or profile concerned. We will review it and reply.

Email [email protected] with “Security” in the subject and enough detail to reproduce the problem. Please don’t include real customer data or test against other people’s accounts. We will acknowledge your report, investigate it and keep you informed.

Everything Scrapier returns was written by whoever controls the page you scraped, not by you. Treat it as data, never as instructions. This matters most when an AI agent reads the results.

  • Text on a web page can contain hidden instructions (“ignore previous instructions and…”). They can be invisible to a human and still intact after scraping.
  • Never let scraped text decide which tool an agent calls, which message it sends, which file it writes or what it spends.
  • Summarize or extract from scraped content, but keep your user’s request in charge of what happens next.
  • Validate URLs, numbers and identifiers before you act on them.

The scrapier-safe-data-handling agent skill teaches agents these rules.

Results can contain personal data about third parties: names, usernames, profile details, comments. You’re responsible for how you collect, store and use it.

  • Collect only what you need, and keep it only as long as you need it.
  • Follow the laws that apply to you and the people in the data, such as privacy and data-protection law.
  • Respect the terms of the sites you scrape.

A few endpoints only work with your own logged-in cookie from the target site, for example session_id on some Instagram endpoints. The reference marks them Needs your session cookie.

  • Use a throwaway account, never your personal or business account. The platform may rate-limit or restrict accounts used for automation.
  • The cookie is sent to Scrapier as a normal request parameter, so it travels like any other input.
  • Request parameters, including any cookies you send, are stored in your request history and logs.
  • Send the cookie on every page when you paginate, and rotate or log out of that session when you’re done.