Shlumi is the AI agent firewall that detects Claude, Gemini, GPT-Operator, Playwright, Puppeteer, and Selenium browser automation in real time — scoring every session, enforcing block and challenge rules, and giving you full forensic visibility.
No credit card required · Free tier includes 1,000 sessions/month
Sessions Today
1,284
+12%
Bots Blocked
47
+3
Avg Score
23.4
clean
Detection Rate
94.2%
↑ 2.1%
Features
Built for developers who need real AI bot detection — not just a CAPTCHA that AI agents can solve.
Detect navigator.webdriver, CDP leaks, Playwright globals, Claude extension signals, and 30+ other browser automation markers in real time.
Every session receives a 0–100 confidence score based on weighted signal analysis. Verdicts are clean, suspicious, or bot — with full explainability.
Define custom rules to block, challenge, flag, or throttle sessions based on score thresholds, specific signals, or verdict types.
Fire real-time POST payloads to your own server on every block or challenge event. Integrate with Slack, PagerDuty, or your own firewall.
Every session is automatically enriched with country, ASN, and datacenter/VPN/Tor flags — no extra configuration required.
Embed a single <script> tag on any website. Works with Next.js, React, plain HTML, WordPress, Webflow, and any other stack.
How it works
Three steps from sign-up to blocking AI browser automation on your production site.
Embed the script
Add one line of JavaScript to your website. The script runs silently in the background and collects browser fingerprinting signals.
Embed the script
Add one line of JavaScript to your website. The script runs silently in the background and collects browser fingerprinting signals.
<script src="https://shlumi.com/agentshield.js"
data-key="as_live_your_api_key">
</script>Signals are scored
Each session's signals are sent to the Shlumi API, scored against 30+ detection rules, and assigned a confidence verdict.
Signals are scored
Each session's signals are sent to the Shlumi API, scored against 30+ detection rules, and assigned a confidence verdict.
// Server-side middleware check
const res = await fetch('/api/shield/check?sessionId=' + id);
const { action, verdict, score } = await res.json();
if (action === 'block') return res.status(403).end();Enforce in real time
Your server-side middleware queries the /api/shield/check endpoint to enforce block, challenge, or throttle decisions before serving content.
Enforce in real time
Your server-side middleware queries the /api/shield/check endpoint to enforce block, challenge, or throttle decisions before serving content.
// Response from Shlumi
{
"action": "block",
"verdict": "bot",
"score": 94,
"signals": ["webdriver", "cdp_leak", "claude_extension"]
}Pricing
Start free. Upgrade when you need more sessions or advanced features.
Free
For personal projects and experimentation.
Pro
For production websites and growing teams.
Business
For high-traffic sites and security teams.
Join developers who use Shlumi to detect and restrict AI agent automation before it scrapes your content, abuses your APIs, or bypasses your paywalls.
No credit card required · 1,000 sessions free every month