Pluno - Browser agent that’s 10x faster than Claude
by•
Pluno just killed Claude in the browser. Claude clicks around the UI like a drunk intern - instead, Pluno talks directly to the API to act 10x faster, with 10x fewer tokens. Learns new tools instantly and uses every web app like a pro. Tested and outperformed Claude in 500+ web apps. Claude clicks - Pluno gets it done. Launch day: get $50 in free credits.
Replies
Pluno
Hey makers!
Here’s what Claude & other browser agents got wrong:
They use software like humans.
They wait for pages.
They stare at screenshots.
They click buttons.
They get stuck in dropdowns.
You’ve probably seen this - impressive for a demo, painful when you actually need work done.
So we built Pluno.
Pluno skips the UI entirely and talks directly to the APIs underneath your web apps. You tell it the outcome, and it gets the task done in the background.
No clicks.
No dropdowns.
No babysitting.
Just: Done ✅
You can use Pluno as a browser extension to update records, extract data, fill forms, run bulk actions, create dashboards, and handle annoying SaaS tasks that normally eat up your day.
We benchmarked Pluno across 312 real-world tasks in 24 tools (incl. HubSpot, Notion, Stripe, Linear), and the results speak for themselves: 34% higher success rate than Claude’s browser extension, at ~14x the speed.
Why we’ve built it? Because we loved the idea of Claude’s browser extension, but in reality it was way too slow to be useful.
For founders, operators, and tiny teams doing the work of ten people:
This one is for you 💜
What I’d love to know: Which browser task would you use to test Pluno’s limits?
P.S. Get 50$ in free credits, today only.
@syed_ali319 Seems interesting, will it be able to bypass CAPTCHAs?
Pluno
Hi @chatabhishek - it can't bypass CAPTCHAs currently.
But it will use your user session, so you can login once and then let Pluno handle the tasks inside.
Which use case did you have in mind where you need the CAPTCHA bypass?
@korbinian_abstreiter Many websites ask users to complete a CAPTCHA before filling out a form. Common examples include appointment booking and job applications.
Pluno
@chatabhishek totally makes sense, thanks for sharing!
Would love to hear your feedback how it's working for you :)
Pluno
Korbi here - Technical Co-Founder of Pluno.
Here's why we've built Pluno, and what browser agents got wrong:
Imagine an AI voice agent talking to another voice AI to schedule a barber appointment.
Instead of a 3 second interaction of checking which appointments are available and picking one, they'll converse back and forth for 10 minutes.
Same is happening for browser agents using the browser.
They download the UI, make screenshots to see what's visible, and try to reason about clicking the right button.
Instead, they should do what agents do best: Use code, and skip the UI.
Pluno learns the API for each tool you use it on, and can talk directly to the API. It's like having a perfectly designed MCP server for every web app you use.
Try it on the hardest task you can imagine, and let me know if you find one Pluno can't handle - I'll make sure to hook you up with some extra credits in exchange ;)
Is Pluno mainly focused on faster prompting and responses, or does the workflow automation / AI agents side mean it can also take actions across tools?
Pluno
@crystalmei great question! Pluno can take actions across tools, e.g. you could export attendee data from a conference website and directly have them imported into Salesforce.
What use case do you have in mind for taking actions across tools?
Pluno
Been using it internally a lot for creating dashboards on @PostHog, debugging issues with my queries on @Retool, drafting @LinkedIn comments, and exporting data from @leadfeederapp.
What's the most complex task in browser apps you're still doing manually today?
huge fan of the browser plugin. i have a 2nd chrome instance always running where it's doing all my chores. surprised at how good it is!
Pluno
@0xkkonrad Happy to hear that!
If you have any ideas on how we can improve it further, let us know :)
For browser agents, speed only matters if the failure state is inspectable. The handoff should leave behind what it clicked, what it inferred, where it got stuck, and what still needs human approval; otherwise a fast agent just creates faster uncertainty.
Pluno
@krekeltronics That's right. A browser agent should not disappear after a failed run. The durable output is the audit trail: actions, assumptions, blockers, and approvals needed.
The human-in-the-loop feedback cycle only works efficiently if the browser agents acts fast enough, the results are presented well and the AI learns quickly from the feedback
Just installed the chrome extension and I am already a huge fan! Handled an error-case on a website much faster with pluno instead of taking screenshots and uploading them to Claude. Talking directly to the API makes much more sense. Thanks team pluno for the launch and the increase of efficiency you bring with it
Pluno
@timo_dannenmann That's great to hear. Hope you're pushing Pluno to the limits!
The talk-to-the-API-directly bet is the right one for speed, but the wall we keep hitting is that a lot of modern web apps don't expose a callable API in any stable sense. The one I automate safelists its persisted GraphQL queries and 500s anything that isn't on the list, so we're forced back to driving real DOM clicks. How does Pluno handle apps where the underlying calls are signed per-session or the operation set is locked to a safelist?
Pluno
Hey @dipankar_sarkar - Korbi here, technical co-founder of Pluno.
Amazing question, that's exactly why existing browser agents fail on such apps.
Instead, Pluno checks the code of the UI to understand how it makes requests to the server.
Whichever way it is, Pluno learns this functionality and starts communicating to the backend the same way the UI does.
Happy to go deeper on anything you'd like to know!
Skipping the UI and going straight to the API feels like the real shift here - not just a speed improvement, but a completely different way of getting work done.
Curious how you handle actions that aren’t easily reversible though. Is there some kind of confirmation or preview layer, or does it just execute once the intent is clear?
Either way, this is a really interesting direction - congrats on the launch!