Back to home
    Free ToolRuns in your browser — nothing is uploaded

    See what your team actually asks AI to do.

    Drop a Claude, ChatGPT, Gemini or Claude Code export. In seconds you get a map of your AI use — which domains it touches, what data it sees, where the EU AI Act starts to apply.

    Scan your AI chat history

    Export your conversations from Claude, ChatGPT, Gemini or Claude Code and drop the file here. We read what you asked AI to do — here, in your browser.

    Your export file stays in this browser tab — it is never uploaded. Nothing is saved unless you choose to: then only a summary — the platform, the months of first and last use, category counts, model families, prompts per month and an estimated cost. No conversation text, ever.

    How do I export my history?

    • Claude: Settings → Privacy → Export data. Newer exports give you a manifest plus per-category downloads — upload the conversations-000.zip (or conversations.json inside it).
    • ChatGPT: Settings → Data controls → Export data. You receive a .zip by email.
    • Gemini: takeout.google.com → My Activity → Gemini Apps → JSON format.
    • Claude Code: upload a session .jsonl from ~/.claude/projects/.

    What you get, and how it works

    A free tool that turns a chat-history export from Claude, ChatGPT or Gemini into a first inventory of how your team actually uses AI. You drop the export file; the scan counts conversations, groups them over time, and maps the patterns to EU AI Act-relevant categories, so you can see which uses deserve a closer look.

    The file is parsed locally in your browser: chat text never leaves your device — only derived counts are sent, and only if you choose to save the result. From there you can hand the profile to the readiness self-assessment, or save the summary with an e-mail address to start monitoring. A daily scan cap applies per browser. It is a first inventory step, not legal advice.

    Frequently asked questions

    Is my chat history uploaded?
    No. The export file is read in your browser. The scan derives counts and category signals from it; the chat text itself never leaves your device. Only if you choose to save and monitor are the derived counts and your e-mail address sent to us.
    Which exports does the scan read?
    Chat-history exports from Claude, ChatGPT and Gemini. You can also run it on the built-in sample file first, to see what the output looks like without uploading anything of your own.
    What does the scan tell me under the EU AI Act?
    It groups your team's real AI usage into categories relevant to the EU AI Act, so you know where to run the readiness self-assessment first. It is an inventory aid: it does not classify systems and it is not legal advice.

    Cite this tool: Cite or link this tool as «AI Usage Scan», LandingRed — https://landingred.com/ai-usage-scan