Detect fake accounts, coordinated manipulation, and abusive automation. Automatically respond to threats without getting in the way of legitimate users.
Get started →Six connected accounts repeatedly promoting the same products.
Two calls: tell us what happened, and ask us before something risky happens.
npm install @puppetwatch/sdkpuppetwatch.identify(currentUser.id);
puppetwatch.capture("analysis_requested", {
target_type: "ai_inference",
target_id: demo.id,
});const decision = await puppetwatch.guard(currentUser.id, {
targetType: "analysis_requested",
});
if (decision.action === "block") {
// deny the request
}capture() never blocks anything on its own. guard() is what lets you actually stop a bad actor. Call it right before the action happens, and act on what it returns (allow, block, require a CAPTCHA, and more).
Send pseudonymous IDs and event metadata. No emails or message content required.
Integration documentation →Attackers can imitate normal traffic to evade detection. Legitimate automation may have an incomplete or unfamiliar signature and be blocked despite doing nothing harmful.
PuppetWatch monitors what users and agents do on your platform in real time. It detects abusive patterns as they emerge and responds within your limits. A convincing signature cannot erase the activity behind it.
Accounts post fabricated customer experiences to inflate a product’s rating or damage a competitor.
A network of accounts builds credibility, starts discussions, and reinforces each other’s claims to make a coordinated narrative look like organic community consensus. The accounts may play different roles and act at different times. The pattern emerges from how they support the same campaign across discussions.
Accounts systematically collect your catalogue or content, spreading requests across identities to evade limits.
One operator creates new accounts to repeatedly claim credits or introductory offers.
Accounts repeatedly exploit AI inference, exports, or other costly actions.
Stop fake registrations, abusive scraping, and automated exploitation without disrupting useful AI agents and legitimate bots. You decide which automation is welcome.
Fake engagement, bot swarms, and networks of connected accounts can manipulate your platform without any individual action looking suspicious.
Protect your platform without adding friction for everyone. PuppetWatch automatically chooses a proportionate response to suspicious activity.
You set the limits for each action. Choose which restrictions PuppetWatch may apply to registrations, reviews, or other actions on your platform. It selects the response within those limits.
Start safe. Begin in monitor-only mode and review every decision before turning on enforcement.