Tensai - AI Context Bridge

Tensai keeps Claude and ChatGPT in sync while you work across both. Most AI-heavy work doesn't happen in one model - you end up with two tabs open, manually re-explaining decisions, pasting summaries, reconstructing context the other model never saw. Tensai removes that copy-paste loop. HOW IT WORKS Pair a Claude tab with a ChatGPT tab from the Tensai popup. As you work, Tensai captures each exchange and compresses it locally - objective, decisions, open questions, code, and recent context. Switch tabs, and the other AI's draft box is automatically pre-filled with a structured handoff. You read it, edit it if needed, and send it yourself. Tensai fills the box - it never sends anything on your behalf. WHY IT'S DIFFERENT - Local-first - all context stays in your browser's session storage. Nothing is sent to a server, because there is no server. - No account, no API key, no signup - works the moment you install it. - Algorithmic compression - no LLM calls, no added latency, no cost. - You're always in control - every handoff is visible and editable before it sends. Nothing happens silently. GOOD FOR - Getting a second opinion - think through architecture in Claude, then switch to ChatGPT for critique, without re-explaining the setup. Long sessions - when a Claude conversation is approaching its limit, Tensai surfaces a handoff so you can continue in ChatGPT instead of starting over. - Debugging - hand your current state to the other model mid-session and see if a different approach surfaces. - Drafting and editing across models without losing the thread. CURRENT SCOPE Claude ↔ ChatGPT, one active pair at a time. Gemini support and additional providers are on the roadmap. Tensai requests only the permissions it needs to function: storage, tabs, scripting, and alarms. No broader access, no telemetry, no analytics.