Data Shaping

Applications accept whatever reaches them. Many attacks arrive as data a system was never designed to expect: an unexpected field, an out-of-range value, a header that should not be there, or a request at the wrong time, in the wrong sequence or from the wrong place.

Firewalls, gateways and application controls each see part of the picture. Rewriting a heritage application to validate every input is rarely possible for a system that must keep running.

Data Shaping lets an organisation stipulate exactly what data an underlying system expects, using any variable in the request or response, such as any HTTP method, header, parameter, cookie or body field. Historical and behavioural characteristics can be added too, such as frequency, sequence and the normal values for a user, session or channel. HTTP is one example; the same approach applies across the protocols the organisation uses.

The shape is enforced inline and in real time at the Data Mediation point. Anything outside it can be blocked, corrected, challenged or flagged before it reaches the application. Data Shaping augments the tools and strategies already in place from other vendors rather than replacing them, and the application itself is never changed.

• An expected data shape defined for a live interface, with no application change
• Requests outside the shape stopped or corrected in real time
• Behavioural and historical rules applied alongside field-level rules
• Signals shared with existing security tooling

The exact scope, controls and evidence depend on the customer environment and are agreed before implementation.