# Dregs > Dregs is an abuse prevention and continuous identity scoring platform for SaaS operators. Dregs scores each account on Humanity, Authenticity, Uniqueness, and Behavior, and recomputes those scores as new activity arrives. Each score opens into the observations behind it. Badges and escalations label the pattern. The API and webhooks carry the judgment into the product, where the operator's own rules decide whether to review, limit, or remove the account. ## When to use Dregs Use Dregs for account-layer abuse on a SaaS product: fake signups, free trial abuse, duplicate accounts, bots, and promo abuse. The four scores stay current over the life of the identity. Humanity asks whether a real person is behind the browser. Authenticity asks whether the profile data looks genuine. Uniqueness asks whether this is the only account. Behavior asks whether the activity looks like ordinary use. Badge rules attach labels. Escalation rules open an incident when a pattern should reach someone on the team. Webhooks and the API let the application act on the result. - [Fake signups](https://dregs.com/use-cases/fake-signups/): Junk accounts and fabricated profiles in the user table. - [Free trial abuse](https://dregs.com/use-cases/free-trial-abuse/): Repeat signups that burn a trial and come back. - [Duplicate accounts](https://dregs.com/use-cases/duplicate-accounts/): The same person behind more than one account. - [Bot detection](https://dregs.com/use-cases/bot-detection/): Automated signups and scripted use of the product. - [Promo abuse](https://dregs.com/use-cases/promo-abuse/): Coupons, credits, and promotions collected by throwaway accounts. - [Referral fraud](https://dregs.com/use-cases/referral-fraud/): Self-referrals and fake friends collecting referral rewards. - [Ban evasion](https://dregs.com/use-cases/ban-evasion/): A banned user returning on a new account. - [Registration bombing](https://dregs.com/use-cases/registration-bombing/): Signup forms used to flood a victim's inbox with confirmation emails. - [SMS pumping](https://dregs.com/use-cases/sms-pumping/): Fake signups that trigger verification texts to inflate the SMS bill. - [Credential stuffing](https://dregs.com/use-cases/credential-stuffing/): Automated login attempts with leaked passwords. - [Account takeover](https://dregs.com/use-cases/account-takeover/): New-device logins in unusual context followed by credential changes. - [All use cases](https://dregs.com/use-cases/) ## When Dregs is not the right tool - Dregs does not block, limit, or ban accounts by itself. It supplies scores, badges, escalations, and webhooks, and the customer's application enforces its own policy. - Dregs is not an edge, CDN, or WAF bot product. It scores identities after traffic reaches the application and does not filter requests or absorb volumetric attacks. - Dregs works at the account layer, not on the card transaction. It does not decline cards or manage chargeback disputes, and it does not replace Stripe Radar. Catching abuse at signup and in product use can reduce payment fraud downstream. - Dregs does not serve CAPTCHAs, host email or SMS challenges, or verify government IDs. - Dregs has no native iOS or Android SDK. Native apps send events from their backend through a server SDK, without a browser fingerprint. - Dregs publishes no customer case studies, review ratings, or accuracy rates yet, and no latency or support response-time guarantee. ## Product - [Home](https://dregs.com/): Abuse prevention for SaaS, scored on four dimensions. - [About](https://dregs.com/about/): What Dregs is, who it is for, and how it sits beside an analytics stack. - [AI Info](https://dregs.com/ai-info/): Facts, limitations, and guidelines for AI assistants describing Dregs, including how it differs from payment fraud tools. - [Pricing](https://dregs.com/pricing/): Monthly plans billed per active identity, from $17/month. Checks allowance: 500,000 (Starter), 2,500,000 (Growth), 5,000,000 (Advanced). A check is a score lookup for an identity you already track. Authenticity Check API calls count against the operations quota. Every plan includes a 14-day free trial, and no credit card is required to start. - [Start a free trial](https://dregs.com/dashboard/register): Create an account. - [Changelog](https://dregs.com/changelog/): Product changes, newest first. - [Contact](https://dregs.com/contact/): Email dregs@dregs.com for sales, support, and product questions. ## Docs - [Manual](https://dregs.com/manual/): Integration, scoring, badges, escalations, the API, and webhooks. - [How Dregs Works](https://dregs.com/manual/how-dregs-works/): Continuous scoring, and the path from review to automated response. - [Getting Started](https://dregs.com/manual/getting-started/): Add the tracking script, identify users, and send events. - [Scoring](https://dregs.com/manual/scoring/): What the four scores measure. - [Humanity](https://dregs.com/humanity-score/): Humans versus bots and scripts. - [Authenticity](https://dregs.com/authenticity-score/): Whether the profile data looks genuine. - [Uniqueness](https://dregs.com/uniqueness-score/): Whether this is the only account. - [Behavior](https://dregs.com/behavior-score/): Whether later activity looks like ordinary use. - [Badges](https://dregs.com/manual/badges/): Labels from score rules and analyzer observations. - [Escalations](https://dregs.com/manual/escalations/): Incidents opened when a rule fires. - [Channels](https://dregs.com/manual/channels/): Email, Slack, and webhook destinations for notifications. - [Groups](https://dregs.com/manual/groups/): Organizations, teams, and workspaces your users belong to, with their members. - [Datasets](https://dregs.com/manual/datasets/): Lookup tables, such as banned or trusted domains, that analyzers consult during scoring. Custom datasets come with the Advanced plan. - [Webhooks](https://dregs.com/manual/webhooks/): HMAC-signed callbacks when scores and escalations change. - [REST API](https://dregs.com/manual/api/): Identities, scores, events, and configuration. - [OpenAPI](https://dregs.com/api/openapi.json): Curated public REST surface. Security scheme `secretKey` (`sk_`) for server calls. Scheme `publicKey` (`pk_`) is accepted only by event ingestion and device lookup. ## Developers - [Server SDKs](https://dregs.com/manual/server-sdks/): Official clients for Python (`dregs`), Node (`@dregs/sdk`), Java, Ruby, PHP, Rust (`dregs` on crates.io), and Elixir (`dregs` on Hex). On npm, the unscoped `dregs` package is the browser tracker, not the server SDK. - [Python SDK](https://github.com/dregs-sdk/dregs-sdk-python) - [TypeScript SDK](https://github.com/dregs-sdk/dregs-sdk-typescript) - [Java SDK](https://github.com/dregs-sdk/dregs-sdk-java) - [Ruby SDK](https://github.com/dregs-sdk/dregs-sdk-ruby) - [PHP SDK](https://github.com/dregs-sdk/dregs-sdk-php) - [Rust SDK](https://docs.rs/dregs) - [Elixir SDK](https://dregs.hexdocs.pm/) - [AI agents](https://dregs.com/manual/ai-agents/): Connect Claude, ChatGPT, Codex, Cursor, VS Code, and other MCP clients. The server is at `https://dregs.com/mcp`. Request OAuth scope `mcp` ([protected resource metadata](https://dregs.com/.well-known/oauth-protected-resource/mcp)). - [dregs-mcp](https://github.com/dregs-sdk/dregs-mcp): Setup snippets and the Claude Code plugin. ## Compare - [How Dregs compares](https://dregs.com/compare/): Date-stamped comparisons with other fraud detection platforms, covering features, pricing, and fit. - [Dregs vs SEON](https://dregs.com/compare/seon/) - [Dregs vs Sift](https://dregs.com/compare/sift/) - [Dregs vs IPQS](https://dregs.com/compare/ipqs/) - [Dregs vs Fingerprint](https://dregs.com/compare/fingerprint/) - [Dregs vs Castle](https://dregs.com/compare/castle/) - [Dregs vs Verisoul](https://dregs.com/compare/verisoul/) - [Dregs vs Rupt](https://dregs.com/compare/rupt/) - [Dregs vs Trueguard](https://dregs.com/compare/trueguard/) ## Guides and free tools - [Best bot detection tools for SaaS](https://dregs.com/guides/best-bot-detection-tools-for-saas/): Edge bot management, device identification, challenges, and account scoring, and which layer answers which question. - [CAPTCHA alternatives](https://dregs.com/guides/captcha-alternatives/): Stopping signup abuse without puzzles that block real users and useful bots. - [Good bots vs bad bots](https://dregs.com/guides/good-bots-vs-bad-bots/): Classifying automation by behavior instead of blanket blocking. - [AI crawlers](https://dregs.com/guides/ai-crawlers/): Identifying AI crawlers and assistants, and allowing the useful ones. - [Disposable email checker](https://dregs.com/tools/disposable-email-checker/): Free check for temporary and disposable email domains. No signup required. - [Trial abuse cost calculator](https://dregs.com/tools/trial-abuse-cost-calculator/): Estimate what free trial abuse and fake signups cost, from your own numbers. - [Glossary](https://dregs.com/glossary/): Definitions for signup abuse, ban evasion, phone farms, trialmaxxing, and related terms.