Skip to content

Latest commit

 

History

History
52 lines (39 loc) · 2.42 KB

File metadata and controls

52 lines (39 loc) · 2.42 KB

Help me learn scapy?

What is all this?

We're working with insulaudit, a python library to audit and interrogate medical devices. The protocols that govern our insulin pumps are locked, ensuring incorrect therapy without recourse. Meanwhile adverse events are attributed to "user-error."

Now that I'm a bit more familiar with how the basic model works, I was hoping to use scapy to round out the corners. see decoder.py

But I could use a few pointers in the right direction. I've been reading through some other protocols:

Plus reading through scapy sources, the ASN.1, stuff, and looking at how bind_layers works.

It appears to me as if multiple requests are required for one logical fetch of information, and I'm wondering how to build this up in scapy.

One kind of mega-model that attempts to sniff for context, or is there a clever way to do this with thin wrappers that bind correctly depending on the context?

It's important to note, I can identify most of the bytes involved here:

The big problem is that we cannot interpret the actual contents from the payloads. However, the captures in decoding-carelink are 100% under my control, with data entered by me, constrained to a single day. And we have made some progress in identifying some of the contents: php analysis

So the task immediately before me, I suppose is to patch the tool so that it produces packets of command requests and the payload contents for further analysis?

Any pointers, help patching MUCH appreciated.