Launched this week
Lyto
"One AI agent across your browser, tools, and messages "
272 followers
"One AI agent across your browser, tools, and messages "
272 followers
Lyto AI is a Chrome extension that gives you full control over your browser. Open and close tabs, scroll, click, fill forms, and interact with every DOM element. Integrates with Google Docs, Gmail, and Google Sheets. Research, automate tasks, and organize your workflow — all inside Chrome.
Free Options
Launch Team / Built With
@luki_notlowkey Agreed, it's the hard part. Two layers: a tiny always-on core of who-you-are, and everything else as facts pulled in on demand only when relevant, so the agent isn't dragging its whole history into every message.
The part I like most is that it consolidates in the background, a bit like sleep. On a schedule it re-derives the stable picture of you and quietly drops the duplicates, the one-off requests, the noise that won't matter next week. So "what to forget" isn't decided in the moment, it settles over time, which is the only way I've found that actually scales.
The 'across browser, tools, and messages' pitch raises an obvious question for me - how much context is being sent to the server when it bridges across those apps? Genuinely curious about the privacy model here before I'd feel comfortable giving it message access.
@omri_ben_shoham1 Fair thing to ask before granting message access. Lyto runs the agent server-side, so context for a given task does reach the server, but it pulls what a task needs on demand (a specific email, the current page) rather than continuously syncing or mirroring your whole inbox. The conversation with Lyto itself is kept so it has memory across sessions, and credentials are stored encrypted at rest. We don't train on your messages, and our model provider is on a paid tier where they don't either. Happy to get specific on any part of that if it'd help you decide.
@gleb_babichev appreciate the detail, "on demand, not continuously synced" is the part that actually matters to me. still, "encrypted at rest" covers storage but not the model provider's paid-tier retention window - do you know the actual retention period there, or is it indefinite until you delete the conversation?
Good distinction, and it's not indefinite on the provider side. Our model provider is Google (Gemini API, paid tier): on paid they don't use prompts or responses for training, and requests are logged for up to 55 days purely for abuse/safety detection and legal compliance, then deleted automatically. There's also a Zero Data Retention mode we can request per-project, where content and identifying metadata get stripped before anything is logged at all, so we're looking at that.
That's separate from the Lyto conversation itself, which we store so the agent keeps memory across sessions. That one is retained until you delete it, either the single conversation or your whole account, which triggers a full cascade erasure. So neither side is "indefinite by default": the provider window is bounded at 55 days, and your own history is deletable on demand.
appreciate the detailed breakdown, that's more transparency than most agent products give on this. the cascade erasure part is the one i'd actually want to test myself before trusting it fully, deletion claims are easy to write down and harder to verify from the outside
That’s a fantastic idea!
Did the inspiration come from a tedious task you personally had to deal with?
Or was it based on requests from users?
Lyto
@nanimono_masa Both honestly! I kept losing context every time I switched between tabs and had to re-explain everything to ChatGPT from scratch. That was the personal frustration. Then when we talked to other people we realised it was not just me, it was everyone bouncing between tools all day. That combo is what pushed us to build it.
@arystan_tanekov I see—it certainly is a tedious task.
Thank you for the answer.
Hey, how does Lyto handle the case where the source data or target workflow is inconsistent instead of clean?
Lyto
@romejerome Great question. When the data or workflow is messy, Lyto flags the inconsistency and asks for clarification before proceeding rather than guessing and getting it wrong. The goal is to never silently do the wrong thing. Happy to go deeper on a specific case if you have one in mind!
Mailwarm
How do you handle safety, like do you require confirmation before it submits forms or sends emails in Gmail?
@thamibenjelloun Yes for the irreversible stuff. Sending an email, submitting something you can't take back, anything destructive or financial, it shows you exactly what it's about to do and waits for your go-ahead. Reversible edits (typing into a field, a doc change you can undo) just happen, so it doesn't nag you on every keystroke. The line is "can this be undone", and confirmation sits on the side that can't.
Good to hear the gate is on mutative actions. The case that bit me building browser automation: a Gmail send is easy to intercept because there is a discrete click to gate, but Sheets and Docs commit on blur, there is no submit event. So a cell edit is already live by the time you would show a preview. Do you stage those edits before writing back, or is the confirm only on actions with an explicit send button?
One agent that follows me across the browser, my tools, and my messages is a smart angle, most assistants make me come to them in one tab.
Lyto
@yibo_wang3 Exactly, most AI makes you go to it. Lyto comes with you. That shift is the whole thing.