They both have X in their names, just imagine the synergies!
189 karma · joined December 19, 2024
Our own DOM-only web agent, rtrvr.ai, worked seamlessly underneath dialogs, can just use off the shelf Gemini Flash Lite and use Chrome native APIs leading to minimal infrastructure failures, SOTA performance and lowest cost.
We essentially leveraged sandboxes built into Chromium browsers for LLM generated code execution.
This actually simplifies a lot of the setup in the blog post, as it leverages existing sandboxing infra exposed for extensions: https://developer.chrome.com/docs/extensions/how-to/security...
Chrome Extension can also access local files and can also execute LLM generated code in sandboxes
Sounds like a skill issue, our web agent is able to interact with cross origin iframes to for example solve captchas: https://www.youtube.com/watch?v=LD3afouKPYc
We honestly haven't faced any bot detection or blocking issues. Owning the browser layer exposes to you much more detection just look at Comet getting blocked on Amazon etc.
I still don't buy the we needed it to be a whole Browser and not a Chrome Extension argument:
- your interface is still literally a chrome extension side panel
- none of the agentic browsers from the bigger players like Atlas and Comet really took off either
I do think the server side integration is required:
- with rtrvr.ai a ton of users are integrating our web agent chrome extension via Remote MCP from chatgpt.com as well as triggering as an API endpoint remotely. Your implementation is limited to only local connections as I understand.
- the biggest unlock for users is running at scale, so just being able to launch a hundred cloud browsers, do a task, and return results while you do other things. So we see hybrid cloud/local execution as the key unlock for this year
Your workflow pipeline is really cool! Any blog post/summary on how you set it up?
Last year was a lot of technical builders exploring the capabilities, and I am excited for this year of making these agentic browsers useful!
I will give it a couple of days for them to tweek it back
You can easily bring up bribing and retaliation as excuses why we shouldn't have jury trials either. These were never really fundamental problems with open democracy, like Andrew Jackson didn't bribe everyone in the country to become president.
TBH if a politician offered me $100 to listen to his pitch, I would take it but I would still vote based on the thousands of dollars of lifetime impact of their policies on my income and assets.
- its free
- its remote, so no local resource use and persists
- user can check results after as well as share
Quick demo: https://www.youtube.com/watch?v=Sas8iWMCBA4
This article is right about secret internet voting: it’s fundamentally incompatible with unsupervised devices and global networks. But secrecy is the constraint that breaks everything.
If you instead require public, verifiable voting, most of the "unsolved" problems disappear. The core requirement becomes: everyone can independently verify inclusion and correct tallying.
That’s where blockchains are a genuine game-changer: - They provide a public, append-only, tamper-evident system of record.
- Anyone can recompute the tally from first principles — no trusted servers, no “checker apps,” no special dispute resolution.
- Server compromise or insider attacks stop being catastrophic; fraud becomes immediately visible rather than silently scalable.
- Malware can still affect an individual’s vote, but it can’t secretly change the election at scale — the main failure mode highlighted in this post.
If trust is the goal, opacity is the wrong primitive. The secret ballot is mistaken path solving a non existent and purely theoretical problem of vote buying.
In a world where we expect everything to be easily accessible, the hardships placed by all the steps required to vote (registration, confirming residency location, waiting in line for polling booth) is seriously impacting voter participation. We need to get with the times and modernize this voting infrastructure.
Goal is to raise funding and then fill back the vowels
We only request drive.file permission so create new sheets or access to ones explicitly granted access to us via Google Drive Picker
Working on a competing extension, rtrvr.ai, but we are more focused on vibe scraping use cases. We engineered ours to avoid these sensitive/risky permissions and Claude should too, especially when releasing for end consumers
We actually created shortcut for the agent itself to go to AI Studio, create API key, and configure itself haha: https://www.youtube.com/watch?v=L6ywfLrG_W8
If you're trying to figure out how to track product visibility/rankings on ChatGPT without manually typing queries 50 times a day, check out this new tool: rtrvr ai!
The problem is that standard scrapers usually get blocked by OpenAI/Perplexity, and using the official API doesn’t give you the "Web Search" results (citations, sources, UI elements) that a real consumer sees.
You can get around this with rtrvr ai by turning your own Chrome Browser into an API endpoint.
We give a copy/pasteable MCP url that you can use with your favorite agent/chatbot/site and give those providers browser context and allow them to do browser actions.
So compared to Playwright MCP and others that require you to run npx and can only be connected to local clients, with ours you just paste a url and can use with any client.
Checkout our recent posts: https://news.ycombinator.com/item?id=45898043 https://www.youtube.com/watch?v=B4BTWNTuE-s
I get it no one is using that, but like this just sounds like a rehash?
https://modelcontextprotocol.io/specification/2025-06-18/ser... https://modelcontextprotocol.io/specification/2025-06-18/ser...
But if a website changes its UX and your recording no longer works then it will fail?
Working in the browser agent space myself, although you save on cost with these repeatable recordings the true disruption of browser agents is using one prompt on thousands of websites without having to worry about maintenance at all
As an aside, we are about to launch something like similar at rtrvr.ai but having AI Web Agents navigate pages, fill forms and retrieve data. We are able to get our costs down to negligible by doing headless, serverless browsers and our own grounds up DOM construction/actuation (so no FireCrawl costs). https://www.youtube.com/watch?v=gIU3K4E8pyw
Can only imagine the pitch, yes please give us billions of dollars. We are going to make a huge investment like paying of our lawsuits.
TLDR:
- Using VERY HIGH RISK Debugger Permission that malicious websites can exploit to get device access. Very surprising a major tech company shipping product with such risky permissions to consumers. More info on debugger risks: https://dspace.networks.imdea.org/bitstream/handle/20.500.12..., https://issues.chromium.org/issues/40091993.
- Prompt injection risks combined with Debugger permission on user device is asking for trouble.
- Will trigger captchas/bot detection even on your normal browsing due to this permission.
- Kind of slow. Limited to current open tab as opposed to capability of multi tab action because only current active tab get rendered. For example rtrvr.ai can open a batch of tabs and take actions on background tabs.
- For some websites like Bloomberg asking to go to claude.com
- Using VERY HIGH RISK Debugger Permission that malicious websites can exploit to get device access. Very surprising a major tech company shipping product with such risky permissions to consumers. More info on debugger risks: https://dspace.networks.imdea.org/bitstream/handle/20.500.12..., https://issues.chromium.org/issues/40091993.
- Prompt injection risks combined with Debugger permission on user device is asking for trouble.
- Will trigger captchas/bot detection even on your normal browsing due to this permission.
- Kind of slow. Limited to current open tab as opposed to capability of multi tab action because only current active tab get rendered. For example rtrvr.ai can open a batch of tabs and take actions on background tabs.
- For some websites like Bloomberg asking to go to claude.com
Most of the other agentic chrome extensions so far used vision approach and sensitive debugger permissions, so unsure if Anthropic just repackaged their CUA model into an Extension.
- an AI Web Agent that autonomously completes tasks, creates datasets from web research, and integrates any APIs/MCPs – with just prompting and in your browser!
As an ex-Google engineer I know the immense engineering efforts and infrastructure setup to develop Chrome. It is very implausible that two people can handle all the effort to serve a secure browser with 15+ million lines of constantly changing C++ code.
A sandboxxed browser extension is the natural form factor for these agentic capabilities.
Our benchmark results [https://www.rtrvr.ai/blog/web-bench-results] show that we are 7x faster than browser-use so curious to see if your claims live up to the hype