You are deep in a conversation, the model finally understands what you are building, and then the daily limit hits. The usual move is to open another AI and explain everything from scratch. This removes that.
The tool is a free Chrome extension called Promptive Sentry, by a developer called Xenriq Systems. I have been testing it, and the core idea is genuinely useful, but there is a privacy nuance the one-line pitch skips, and I am going to be straight with you about it, because the whole thing hinges on it.
What It Actually Does
The headline feature is Save Context. In an active chat, you click it and the extension reads the entire conversation and packages it into one structured summary bundle. You then open a different assistant, click reload, and it injects that context back in, so the new model picks up roughly where the old one left off. It works across the three big consumer chat sites.
- ✓Save Context and reload across Claude, ChatGPT, and Gemini. This is the reason to install it.
- ✓Token usage tracking shown directly on Claude.ai, with rolling usage windows so you can see a limit coming before it lands.
- ✓A one-click prompt enhancer on all three sites, with a few tone modes.
At the time of writing it sits at a 4.6 rating, around 10,000 users, on version 1.1.13, last updated in June 2026. It installs like any extension, and yes, it is free to add.
Chrome Web Store, search "Promptive Sentry" Direct: https://chromewebstore.google.com/detail/promptive-sentry-%E2%80%93-author/ikbkijdgnelcijmdkcaaoabmobagakim
The Part the Pitch Skips: Where Your Chat Goes
The listing describes the tool as "local-first," and that is true for one thing: the saved bundles and your settings are stored on your own device. But the act of creating the summary is not local. This matters, so here is exactly what happens when you click Save Context, taken from the developer's own privacy policy.
Your saved bundles
Once a summary is made, the bundle and your preferences are stored privately in your browser. That part does not leave your machine.
The full transcript, to summarise it
To build the summary, the extension sends your entire conversation over HTTPS to the developer's own backend on Google Cloud, which runs it through Google's Gemini model, then returns the summary.
So your Claude or ChatGPT conversation transits a third party's server and a Google AI model on its way to becoming a portable bundle. The developer states this processing is transient, that the transcript is "never logged, saved to a database, or stored," that there are no accounts and no trackers, and that the only persistent identifier is a random device ID used to rate-limit abuse. Their named infrastructure is Google Cloud Run for hosting, Google Gemini via Vertex AI for the summarising, and Upstash Redis for rate limiting.
"Local-first" describes where the bundle is stored, not where your conversation is read. Those are different claims.
None of this is unusual for a summarisation tool, and there is no evidence of anything shady. But "it reads your whole conversation and sends it to a server" is a materially different thing to agree to than "it works entirely on your device," and you deserve to make that call with the real picture. Two honest flags: the "never stored" promise is a trust-the-policy statement, not something you can technically verify, and the policy names only Google as a processor even though it reads Claude and ChatGPT chats.
So Should You Use It?
- 01Great fit if you are constantly hitting free-tier limits on general work, brainstorming, drafting, and learning, and you want continuity without retyping context.
- 02Think twice for anything confidential: client data under NDA, private code, personal or regulated information. Do not click Save Context on a chat you would not be comfortable sending to a third-party server.
- 03A free built-in alternative exists for the Claude-to-Claude case: Anthropic's own memory and chat search already carry context between your Claude sessions, entirely within Anthropic. The extension's edge is going across providers.
My take: it is a clever, well-made tool that solves a real, daily annoyance, and I am comfortable using it for general work. I just would not run it on a conversation I would not be willing to forward to a stranger's server, because functionally that is what Save Context does. Use it with that line in mind and it is a genuine upgrade to how you move between models.
Want this kind of agentic setup for your own brand or product?
I help founders and brands build generative AI systems that actually ship: content engines, agentic workflows, and AI-built sites that are production-grade down to the schema layer. If you want a second pair of eyes on your setup, book a session.
Book a Growth Chat
Extension details, ratings, and user counts reflect the Chrome Web Store listing for Promptive Sentry as of July 2026. The data-handling description is drawn from the developer's published privacy policy, which names Google Cloud Run, Google Gemini via Vertex AI, and Upstash Redis as processors and states that transcripts are processed transiently and not stored. I have no affiliation with the developer and this post is not sponsored. Verify the current listing and policy yourself before installing: the Chrome Web Store page.