Living Threat Models

[pdxjohnny]

Meta thread do drop notes and scratchwork in around Living Threat Model (LTM) work driven by @johnwhiteman.

[pdxjohnny]

2022-04-18 1:1 John/John - LTM and DFFML

• TODO
  • Andersen to implement caching
    • These we should revisit after we implement the dataflow as class PR
      • Implement LoadSourceInputNetwork
      • Implement UpdateSourceInputNetwork
    • Implement SourceInputNetwork
      • Use source within config since Input IDs are unique
  • Andersen to implement auditor / visualization based on cached workflow
    • Load cached flow with different output operations
  • Andersen to answer question of how do we have structured log output
  • Whiteman to think about what are good open source collectors we should write
    • Go around an grab AST
      • https://docs.python.org/3/library/ast.html
    • Look at AST
      • grep for things that look like inputs

[pdxjohnny]

• Vulnerability Checks
  • Using HTTP vs. HTTPs (detect verify=False)
  • Detect discrepancy between networking not declared and usage of requests
  • Make some code which would be the python which good.json/bad.json represent
  • Call John W's parser and auditor along with above requests checks
  • Create THREATS.md
  • Leverage CITATION.cff stlye here there are overlays for each citation format
    • We can imagine STRIDE, etc. as overlays
• Record demo with https://github.com/faressoft/terminalizer