You know that sinking feeling when you install a ‘free’ tool and it asks for a login? That’s your instinct screaming: ‘This is a trap.’ And for 99% of session recorders, it is. They’re not tools — they’re data harvesters dressed up as utilities. But one developer just proved the cloud is a lie. And he built the proof for free.
The cloud isn’t a feature — it’s a toll booth. Every session recorder you’ve tried uploads your clicks, your network calls, even your mistakes to someone else’s server. Why? Because the real product isn’t the recording — it’s the data. They sell it, train AI on it, or lock basic features behind a subscription. The more you use it, the more they own you.
Meet Mohsen1’s Session Recorder — a Chrome extension that does everything the big boys do, minus the backdoor. It captures clicks, typing, network requests and responses, console errors, screenshots, even your voice narration. Everything stays in your browser’s IndexedDB. No account. No backend. No telemetry. You export a zip file. Period.
If your session recorder requires an account, it’s not a tool — it’s a trap. This one doesn’t even ask for your email. Passwords, auth headers, and tokens are masked at capture time. Your voice is transcribed and placed next to the exact moment you were debugging. You can draw arrows and boxes on screenshots. All events line up in a timeline with timestamps.
Here’s the twist that makes this thing indispensable for the AI era: when you export, you pick a detail level with a live token estimate. So a long debugging session perfectly fits your LLM’s context window. No cloud, no subscription, no data leakage. Just pure, local context for your AI.
Privacy isn’t a feature — it’s the absence of a hidden cost. The cloud was never necessary for the core value of session recording — capturing and replaying. It was only necessary for the business model. Mohsen1 proved that the business model is the enemy of utility. His tool is free, open source, and auditable. You can fork it, inspect it, and trust it.
Before you install another ‘free’ recorder, ask yourself: who’s really paying? If it’s not you, it’s your data. The open source alternative is here. Use it.
FAQ
Q: Why would I trust a one-person open source project over a commercial tool?
A: Open source means you can audit every line of code. Commercial tools have opaque business models — they often monetize your data or lock features behind paywalls. This project stores everything locally, has no backend, and no telemetry. You can also fork it and verify the code yourself.
Q: How does this help me with AI debugging?
A: When you export a session, you get a live token estimate. You can choose a detail level that fits your LLM's context window. Then feed the exported data directly into an AI model for analysis. No need to manually sanitize or reformat logs — it's ready to go.
Q: Isn't the cloud better for collaboration and sharing?
A: For real-time collaboration, cloud services add convenience. But for debugging and AI context, local is superior — you control the data. You can export a zip and share it via any channel. The trade-off is privacy vs. convenience. If you value your data, local wins.