Current testing takes a monolithic approach, combining all the possible scenarios into one large test, and comparing against checked in golden files for the engine frames used to generate the packet capture.

Instead, we aim to split this into separate ITs per protocol, with k3po scripts used to drive the engine and produce the engine frames used to generate the packet capture, then a text-based comparison is used to make diffs more readable should an IT fail.

• http extension & payload
• proxy extension
• grpc extension
• sse extension
• ws extension
• filesystem extension
• mqtt extension & payload
• kafka extension & payload
• amqp extennsion & payload
• tls payload
• cover any other frame edge cases not covered above