Product review
Castle account takeover prevention review
Castle evaluates login and other account activity using device, behavioral and network context. Its Risk API exposes an account takeover score alongside bot and account-abuse scores. The platform fits teams that want an application risk engine with policies and investigation tooling.
Editorial score
Application login-risk API
Free access
5 US dollars monthly API budget; up to 1,000 Risk/Filter calls if used only for those calls
Paid entry
Pro: 200 US dollars monthly API budget; Risk/Filter requests 0.005 US dollars each
Key facts
- Output
- Account Takeover Score, Bot Score and Account Abuse Score Vendor page ↗
- Login integration
- Client request tokens followed by server-side login activity Vendor page ↗
- Budget
- Free 5 US dollars; Pro 200 US dollars monthly API usage Vendor page ↗
- Operation units
- Risk/Filter 0.005 US dollars; IP entity 0.001 US dollars, drawing from the same budget Vendor page ↗
Strengths
- ATO-specific risk score distinguishes the login-risk task from generic bot classification.
- Client request tokens and server-side event calls connect collection with account activity.
- Published shared API budgets make a small Risk API evaluation possible.
Limitations
- The free budget is shared between Risk/Filter requests and IP Intelligence; mixing operations changes the number of available calls.
- Client integration and server-side activity calls are required for a complete login-risk implementation. IP lookup alone supplies a different level of context.
Who it fits
Engineering and fraud teams that want inline login-risk decisions and configurable policies while retaining their current authentication service. Include account events and outcomes so a successful login can be assessed in context.
Use the alternatives by operating need to compare authentication, device/account risk, bot defense and identity exposure.
What the score means
The 8.6 out of 10 score is an editorial assessment under the same 40-criterion, seven-category framework. It weighs takeover detection, signal depth, real-time response, integration, evidence, customer friction and lifecycle fit.
Editor’s Choice is a separate recommendation. Read all criteria and relative weights and evaluate the entitled product on the product flow you protect.
How to evaluate it
Integrate client request tokens, successful and failed login events, logout and sensitive updates. Test the ATO score and selected policies against legitimate new devices, low-volume abuse and missing client evidence. Count the actual mix of API operations.
Free API budget and self-serve account
The Free plan includes 5 US dollars of monthly API usage. At the displayed rate it funds 1,000 Risk/Filter requests if no other billable operations consume the budget.
Open evaluation routeSources and further reading
- Product and detection approach · checked October 23, 2026
- Pricing and access · checked October 23, 2026
- Integration and result documentation · checked October 23, 2026
- Account takeover guidance and technical reading · checked October 23, 2026
FAQ
Does the free plan always provide 1,000 login checks?
That is the equivalent when the 5 US dollar monthly budget is used only for Risk/Filter requests at 0.005 US dollars each. IP Intelligence operations share that budget, so a mixed integration has a different allocation.
Can an IP lookup replace the Risk API?
No. Castle distinguishes IP connection intelligence from the deeper user, device and behavior context of a full Risk/Filter integration. Use the operation that matches the intended control.
Explore checkers, demos and evaluation routes →