Browser Use
Hacker ApprovedThe most recommended of 5 browser agents
Open-source framework where an LLM agent drives a real browser, adapting to changing pages. Critics note API key requirements and per-minute cloud billing.
Open-source framework where an LLM agent drives a real browser, adapting to changing pages. Critics note API key requirements and per-minute cloud billing.
77 people gave an opinion
- recommend it
- 52
- criticize it
- 25
- approval share
- 68%
Recommended in 11 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: not recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it carries the badge
- 52 people recommend itNeeds 10 or more, met.
- 2.1 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 11 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Freezes browser state between each action
“Your tool's method of returning element references is clever and should greatly improve llm handling of the page components”
Comment from 12 Mar 2026 in “Show HN: Open-source browser for AI agents (~90% on Mind2Web)” “the solution of freezing JS execution then capturing fresh state after each action is elegant”
Comment from 12 Mar 2026 in “Show HN: Open-source browser for AI agents (~90% on Mind2Web)” “Freezing the browser at every step is a very good approach.”
Comment from 12 Mar 2026 in “Show HN: Open-source browser for AI agents (~90% on Mind2Web)” LLM loop handles dynamic sites
“The LLM loop approach handles this much better than scripted automation because the agent adapts in real-time”
Comment from 30 Nov 2025 in “Show HN: RowboatX – open-source Claude Code for everyday automations” “The framework's LLM loop approach handles dynamic sites much better than traditional selectors”
Comment from 30 Nov 2025 in “The State of AI Agent Frameworks in 2025”
What people warn about
Each mark is one comment
Needs API key and per-minute billing
“it looks like a commercial product that requires an API key to use even if you run the source”
Comment from 13 Nov 2025 in “Benchmarking leading AI agents against Google reCAPTCHA v2” “Browser-use bills per minute, not per millisecond like lambda does.”
Comment from 17 Jun 2026 in “How We Run Firecracker VMs Inside EC2 and Start Browsers in <1s” Captcha and stealth gaps persist
“Exactly these crappy companies like browser use is causing more captcha etc..”
Comment from 17 Jun 2026 in “How We Run Firecracker VMs Inside EC2 and Start Browsers in <1s” “There's still plenty that Browser-Use could improve in terms of stealthiness.”
Comment from 24 Apr 2026 in “Show HN: Browser Harness – simplest way to give AI control of real browser” Heavy dependencies and vibe-coded feel
“Call me impressed between all that vibe-coded crap nowadays and this vibe-coded masterpiece”
Comment from 12 Mar 2026 in “Show HN: Open-source browser for AI agents (~90% on Mind2Web)” “Same experience with browser-use, it installs litellm as a dependency.”
Comment from 24 Mar 2026 in “LiteLLM Python package compromised by supply-chain attack”
Other browser agents people recommend
The four most recommended other browser agents, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Browser UseThis page | 52 | 25 | 68% | 11 | — | Approved |
| Smooth CLI | 10 | 7 | 59% | 2 | — | — |
| PageAgent | 9 | 7 | 56% | 1 | — | — |
| Rover by rtrvr.ai | 7 | 1 | 88% | 6 | — | — |
| TinyFish Web Agent | 6 | 1 | 86% | 2 | — | — |
What people said
40 of the 88 comments that mention it, newest first. Each links to the original comment.
“agent-browser works fine, and it can use browser use as a provider”
“browser-use is a lot more capable and generally more efficient, and a lot easier to embed in other programs”
“I've found it to be faster than playwright MCP.”
“Browser-use bills per minute, not per millisecond like lambda does.”
“Exactly these crappy companies like browser use is causing more captcha etc..”
“Essentially this creates a browser-use model that runs entirely in your browser (no servers).”
“all the wild ways agents currently hack into the browser (agent-browser”
“all the wild ways agents currently hack into the browser (browser-use”
“Using Claude Code with Vercel's agent-browser for some browser automations in Zoho Books.”
“found agentic direction powerful with browser use plugged into a multimodal”
“to parallelize you could use Browser Use Cloud stealth browsers”
“Browser-use is incredible. Solving captchas via proxy is a wild experience when steering in the browser.”
“There's still plenty that Browser-Use could improve in terms of stealthiness.”
“I've achieved pretty much the same thing using the vercel's agent-browser”
“automated vercel agent-browser testing along the way”
“playwright-cli and agent-browser are much much better”
“Same experience with browser-use, it installs litellm as a dependency.”
“it can be combined with agent-browser to get the best of both worlds”
“agent-browser which wraps playwright, both more token-efficient”
“Claude just needs agent-browser and it can generate deterministic code from it”
“Call me impressed between all that vibe-coded crap nowadays and this vibe-coded masterpiece”
“Maintaining the fork isn't so bad, the core chromium changes are only a few hundred lines”
“the solution of freezing JS execution then capturing fresh state after each action is elegant”
“Architecturally, this is probably the best thing I've seen in open source in a while.”
“sounds like this should give it a more accurate slice of time / snapshot of things”
“Your tool's method of returning element references is clever and should greatly improve llm handling of the page components”
“Freezing the browser at every step is a very good approach.”
“A typical page costs 3x to 10x less tokens depending on the content.”
“is a similar take here, and much better than the playwright MCP”
“There are few agents like browser-use, skyvern etc that may provide this capability.”
“full snapshots are more reliable for long multi-step workflows”
“Playwright MCP or Chrome DevTools MCP or agent-browser CLI or something similar”
“It's very effective at doing pretty much any browser automation task and I have to say that using it with the included skill is pretty seamless.”
“helped a lot over playwright but doesn't completely close the gap”
“We use browser-use for password change automation”
“Agent browser is more lightweight than playwright mcp”
“I've been using browser-use for automated password rotation”
“The LLM loop approach handles this much better than scripted automation because the agent adapts in real-time”
“The framework's LLM loop approach handles dynamic sites much better than traditional selectors”
“it looks like a commercial product that requires an API key to use even if you run the source”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.