namuorg

Hey, creator of Browser MCP here.

1. Yes, the extension uses an anonymous device ID and sends an analytics event when a tool call is used. You can inspect the network traffic to verify that zero personalized or identifying information is sent.

I collect anonymized usage data to get an idea of how often people are using the extension in the same way that websites count visitors. I split my time between many projects and having a sense of how many active users there are is helpful for deciding which ones to focus on.

2. The extension is completely written by me, and I wrote in this GitHub issue why the repo currently only contains the MCP server (in short, I use a monorepo that contains code used by all my extensions and extracting this extension and maintaining multiple monorepos while keeping them in sync would require quite a bit of work): https://github.com/BrowserMCP/mcp/issues/1#issuecomment-2784...

I understand that you're frustrated with the way I've built this project, but there's really nothing nefarious going on here. Cheers!

Added!

https://docs.browsermcp.io/setup-server#node-js

Hey Pavel, this is Namu, the creator of Browser MCP.

You’re right, this is an adaptation of Playwright MCP to automate the user’s local browser as mentioned in the GitHub README and here:

- https://github.com/BrowserMCP/mcp/blob/3e6824de6f36eba7d2d3b...

- https://news.ycombinator.com/item?id=43613905

Thanks for all your work to Playwright and Playwright MCP. I’m a big fan!

(For those not familiar, Pavel is the largest contributor to both Playwright and Playwright MCP: https://github.com/microsoft/playwright/graphs/contributors, https://github.com/microsoft/playwright-mcp/graphs/contribut...)

Can you try again?

There was another comment that mentioned that there's an issue with port killing code on Windows: https://news.ycombinator.com/item?id=43614145

I just published a new version of the @browsermcp/mcp library (version 0.1.1) that handles the error better until I can investigate further so it should hopefully work now if you're using @browsermcp/mcp@latest.

FWIW, Claude Desktop currently has a bug where it tries to start the server twice, which is why the MCP server tries to kill the process from a previous invocation: https://github.com/modelcontextprotocol/servers/issues/812

Thanks for the report and the update! I'd love to hear about what you changed — how can I get in touch? I didn't see anything in your HN profile. Feel free to email me at admin@browsermcp.io

Browser MCP uses the Chrome DevTools Protocol (CDP) to automate the browser so it currently only works for Chromium-based browsers.

Unfortunately, Firefox doesn't expose WebDriver BiDi (the standardized version of CDP) to browser extensions AFAIK (someone please correct me if I'm mistaken!), so I don't think I can support it even if I tried.

Of course, you're sending data to the AI model, but the "private" aspect is contrasting automating using a local browser vs. automating using a remote browser.

When you automate using a remote browser, another service (not the AI model) gets all of the browsing activity and any information you send (e.g. usernames and passwords) that's required for the automation.

With Browser MCP, since you're automating locally, your sensitive data and browser activity (apart from the results of MCP tool calls that's sent to the AI model) stay on your device.

An extension is more user-friendly! I leave Chrome open basically 24/7 and having to create a new Chrome instance via the command line just to use Browser MCP just felt like too high of a barrier.

The Puppeteer MCP server doesn't work well because it requires CSS selectors to interact with elements. It makes up CSS selectors rather than reading the page and generating working selectors.

The Playwright MCP server is great! Currently Browser MCP is largely an adaptation of the Playwright MCP server to use with your actual browser rather than creating a new one each time. This allows you to reuse your existing Chrome profile so that you don't need to log in to each service all over again and avoids bot detection which often triggers when using the fresh browser instances created by Playwright.

I also plan to add other useful tools (e.g. Browser MCP currently supports a tool to get the console logs which is useful for automated debugging) which will likely diverge from the Playwright MCP server features.

Aw yay, glad it was beneficial in helping you bringing Intenty to life!

For anyone looking for something similar for their desktop browser, try out Intention: https://getintention.com

Show HN post from 2020: https://news.ycombinator.com/item?id=22936742

I actually built a Chrome extension that runs Claude computer use if you’d like to try it out! [0] It’s currently awaiting approval in the Chrome Web Store.

After having spent the last several years building a popular Chrome extension for browser automation [1], I was excited to see if LLMs could actually build automations end-to-end based on a high-level description. Unfortunately, they still get confused quite easily so the holy grail has yet to come. Still fun to play around with though!

[0] https://autobrowser.ai/

[1] https://news.ycombinator.com/item?id=29254147

For structured content (e.g. lists of items, simple tables), you really don’t need LLMs.

I recently built a web scraper to automatically work on any website [0] and built the initial version using AI, but I found that using heuristics based on element attributes and positioning ended up being faster, cheaper, and more accurate (no hallucinations!).

For most websites, the non-AI approach works incredibly well so I’d make sure AI is really necessary (e.g. data is unstructured, need to derive or format the output based on the page data) before incorporating it.

[0] https://easyscraper.com

I do this for https://browserflow.app (and the AI version in development at https://browserbot.ai) via the chrome.debugger API: https://developer.chrome.com/docs/extensions/reference/api/d...

You can use Browserflow (Chrome extension for browser automation) to extract saved lists from Google Maps: https://browserflow.app/shared/MWz0bTt7zBkyRMTs

Screen recording: https://www.loom.com/share/5ad91e0347294c7a9b4f0e0b3b2b8544?...

I made Browserflow so feel free to ask any questions / let me know if you run into any issues.

Yup, it covers JS navigation. I spent the last four years building a browser automation tool (https://news.ycombinator.com/item?id=29254147) so I was able to cover a lot of bases with Easy Scraper. Try it out!

Thanks! <3

I built a tool called Browserflow (https://browserflow.app) that lets you automate any task in the browser, including scraping websites.

People love it for its ease-of-use because you can record actions via click-and-point rather than having to manually come up with CSS selectors. It intelligently handles lists, infinite scrolling, pagination, etc. and can run on both your desktop and in the cloud.

Grateful for how much love it received when it launched on HN 8 months ago: https://news.ycombinator.com/item?id=29254147

Try it out and let me know what you think!

For anyone who wants a thorough understanding of how to think about React hooks in the context of things like setInterval, Dan Abramov wrote a great piece breaking it down in detail several years ago: https://overreacted.io/making-setinterval-declarative-with-r...

This post helped hooks "click" for me, and once it did, I've absolutely loved them and now thoroughly enjoy writing custom hooks that greatly simplify my code.

When I requested this back in August, it took 5 days to get the results.

You get a long list of links that each triggers a download. It's a bit annoying to bulk download by clicking each link one by one, so I made a Browserflow flow that clicks all the links and downloads everything automatically: https://browserflow.app/shared/61e979ed-47f4-4c94-b5a5-3ade0...

Thanks Renat! Bardeen looks slick and your team clearly put a ton of work into those integrations. Good stuff!

Not sure what you're referring to with the "input" action, but there's a "Loop Spreadsheet Rows" command that will let you perform a set of actions per row in a spreadsheet.

There aren't immediate plans because it hasn't been a common request, but I'm open to it if there's enough interest!

Feel free to email me at dk@browserflow.app to discuss your use case.

This is simply not accurate, and you can easily test the claim. Just try running Browserflow on the sites you're thinking of and you can see for yourself whether it's instantly banned or not.

Glad to hear it!

Browserflow saves to Google Sheets which can be converted to other file formats as needed using other tools.

That's totally fine. I chose this model so that I can create a sustainable income stream to continually support and improve Browserflow, but you don't have to agree.

FWIW, I did consider the one-time payment and upgrade model, but it's not possible for Chrome extensions because the distribution is controlled by Google and all users are automatically upgraded to the latest version.

Cheers!

Hello, my fellow engineers. :)

The reason for having limitations for local automations is a business one, not a technical one. I like the idea of scaling the price based on how much value Browserflow provides (if it saves someone hours of work, paying less than 20 bucks should be a no-brainer) so that's why there are runtime limits by plan.

Oh wow, you're a real one. Glad to see you here. :)

I know, right? HTML is the ultimate API and I also found it strange that there aren't tools around helping people work with it more effectively. Hence Browserflow!

The biggest challenge has been getting over the "just one more feature..." trap and actually launching the thing. I still see so many areas for improvement, but it's nice to see that people appreciate how far it's come.

In terms of technical challenges, the biggest one by far has been creating flows that are reliable over time. Sites change all the time so having automation that's reliant on HTML structure, CSS classes, or pieces of text is inevitably going to be brittle, but I want to minimize how often flows break. Browserflow aims to generate flows that will remain stable for as long as possible, but there's certainly a lot of room for improvement.

Hey Adam, it does! It was a pain in the butt to build so I'm especially proud of it. You can use the "Upload File" command to attach a file.

If you're running the flow locally, you can put in a local path. If you want to run it in the cloud, you'll need to first download the file and then upload it.

Yup! All the automation is done via CDP and client-side Javascript.