Invite-only beta · 12 spots remaining

Stop reverse-engineering websites. Let an agent do it.

browser-recon is an AI agent for scraping reconnaissance. Browse a site for two minutes — our agent returns a production-ready scraping plan: which library to use, which headers and cookies it needs, the safe rate-limit, and a runnable starter script.

Join the waitlist Read the user guide
~10 min
From browse to a working scraper, end-to-end.
90% accurate
On sites where manual recon would take days.
6 stacks
Tested against Akamai, Cloudflare, PerimeterX, DataDome, Imperva.
Tested against
Walmart
Staples
Target
Airbnb
Ticketmaster
CoinMarketCap
How it works

Three commands. One scraper.

You install the CLI, browse the site for two minutes, and get a complete plan back. No proxies to manage, no anti-bot fingerprints to identify by hand, no trial-and-error scripts.

01 / Install

Drop one command in your terminal

pipx install browser-recon · 128KB, no proprietary code shipped to your machine.

02 / Browse

Show the agent what you want

Chrome opens. You click on what matters — a search, a product, a review page. Press Ctrl+C when done.

03 / Receive

Get a plan that actually works

The agent returns a verified recommendation: library, headers, cookies, rate-limit, and a runnable Python script.

What you get

Recon that's grounded in measurement.

Most scraping tutorials guess at which library will work and which headers matter. browser-recon doesn't guess — it actually fires test requests through real proxies and reports which combinations succeed.

The output isn't advice. It's a working scraper, with the exact library, headers, cookies and timing the live site responds to.

01

Anti-bot fingerprinting

Detects Cloudflare, Akamai, PerimeterX, DataDome, Imperva — and what each one will do to a naive scraper.

02

Real validation, not guesses

Fires test requests through real proxies. Reports which library × proxy combination actually returned the data.

03

Production cost projection

Measured bandwidth × your proxy rate = a real dollar figure per 1,000 requests. Not a vague band.

04

Runnable starter code

A Python script with the recommended library, headers, cookies and delay. Drop it into your stack and it works.

Dashboard

Every scan, every report.

One place for your scan history, credit balance and report URLs. Reports stay live for the duration of your plan.

browser-recon.com/dashboard

Recent scans

New scan
CREDITS
23 of 30 this month
SCANS THIS MONTH
7
PLAN
Pro · renews May 30
walmart.com
curl_cffi · chrome120 · residential
complete 7 / 9 matched $0.84 / 1k View report →
staples.com
requests · datacenter
complete 9 / 9 matched $0.05 / 1k View report →
ticketmaster.com
curl_cffi · safari17 · residential
replay 22% 2 / 9 matched $1.21 / 1k View report →
Reports

Every recommendation. Sourced from real measurement.

Each report opens with the verdict, the recommended library and proxy tier, and a runnable starter script. Drift detection re-checks the recommendation on schedule.

browser-recon.com/r/gpeni1un4v65
VERDICT

walmart.com runs dual Akamai + PerimeterX. curl_cffi with chrome120 + residential proxies reproduces the captured data with cookie warmup.

Estimated cost: $0.40–$2.00 per 1,000 requests using residential proxies. Cookie warmup required before the Reviews GraphQL endpoint will respond. Confidence: 0.82.

Library
curl_cffi
Impersonation
chrome120
Proxy tier
residential
Cost / 1k
$0.84
Pricing

Pay for what you scan. Nothing else.

Credits expire monthly. Reports stay live for the duration of your tier — re-scan for one credit to refresh.

Limited · first month only
Tester
$5 / mo
5 credits
  • 5 scans / month
  • Reports kept 2 weeks
  • Email support
  • One-time signup only
Join waitlist
Beginner
$10 / mo
12 credits
  • 12 scans / month
  • Reports kept 1 month
  • Email support
Join waitlist
Pro Max
$60 / mo
100 credits
  • 100 scans / month
  • Reports kept 3 months
  • Priority queue
  • Slack support
Join waitlist
Need higher volume, a custom report shape, or scraping code generated for your framework? Talk to us about Enterprise →

Get on the list. Skip it with a tweet.

We're letting a small batch in each week. Drop your email and we'll let you know when your turn comes up — or post about us on X and we'll let you in next.

247 developers in line · 12 spots opening this week