3 items
Privacy-first AI usage audit tool for academic compliance. Tracks metadata only — no prompt or response content is ever stored. Automatically log your ChatGPT & Claude usage for academic compliance. Generates a formal AI declaration — no content stored. Many universities now require students to declare their use of AI tools when submitting assignments. AcademiTrace makes this effortless — it runs quietly in the background and automatically builds a verified log of your ChatGPT and Claude interactions. 🔒 PRIVACY-FIRST — NO CONTENT STORED AcademiTrace never reads, stores, or transmits your actual prompts or AI responses. It captures only: • Timestamp of each exchange • Platform used (ChatGPT or Claude) • Word count of prompt and response • Page/conversation title • Date of session All data is stored locally on your device. No servers. No accounts. No tracking. 📋 GENERATE AN ACADEMIC DECLARATION IN ONE CLICK Export a formatted "AI Usage Declaration" document ready to attach to your thesis, essay, or assignment. The report includes a summary, per-platform breakdown, session history, and a signature field — meeting the transparency requirements of most institutional AI policies. ✅ KEY FEATURES • Automatic tracking on chatgpt.com and claude.ai • Supports logged-in and guest (logged-out) sessions • Multilingual word count (English, Chinese, Japanese, Korean, and more) • Period filter: Today / 7 days / Month / All time • Per-site usage breakdown • One-click TXT export with academic declaration template • Two-step confirmation before clearing data • Zero external network requests — fully offline capable 🎓 DESIGNED FOR PhD students, researchers, lecturers, and anyone who uses AI tools as part of their academic or professional work and needs to document that usage transparently. PERMISSIONS USED • storage — to save your usage log locally • Access to chatgpt.com and claude.ai — to detect interactions on these sites only No other permissions are requested.
rating_count is the Chrome Web Store ratings count, not a written-review count.