Skip to main content

Returning to the Challenge Labs (GPO Abuse and Workflow Optimization)

·445 words·3 mins
  • Worked through Challenge Lab 0, Secura, again over the past few days.
  • This is the easiest of the labs and wouldn’t normally take so long to complete, but I’ve also been refining my environment and runbooks to address gaps in my workflow.
  • Unfortunately I won’t be posting writeups for the Challenge Labs, due to OffSec’s copyright and academic integrity policies around PEN-200 content.

GPO Abuse
#

The Secura lab has an opportunity to abuse GPO. I’m trying to avoid depending on Bloodhound as much, so I’m learning to be more hands-on with enumeration and researching exploitation methods.

Workflow/Productivity
#

  • I’ve been using these all for at least a few weeks now, but zoxide, tmux, and direnv have reduced a lot of friction and fatigue when switching between targets in AD labs
  • Pentest-Reference
    • Removed a lot of slop that I never used and could easily look up online
    • Now copying content to a page on website at https://cameroncandau.com/pentest/
    • Established personal conventions for HTTP server and reverse shell ports, to be consistent going forward
    • Documented primary and fallback methods for downloading and uploading files
      • Previously I either had many methods documented, or would just randomly pick a different one from memory each time. Just another way to reduce decision fatigue
  • Added upload functionality to Payload-Server, for quickly transferring files back from a target to my host
  • Uploaded most of the artifacts I’ll need to my ORAS backend to use with Artifact-Locker
    • Having all my scripts and binaries in one place that I can immediately pull down from a fresh Kali VM is awesome. With Payload-Server, it really reduces friction for payload transfers and ensures I’m not scrambling to remember paths or find GitHub repos during the lab.
  • Overall, I’m finally feeling pretty satisfied with my tooling, runbooks, and note templates. I’ll have to detail some more of these projects in future posts.
  • Ending ChatGPT Codex subscription today
    • Feeling a bit uneasy because I think it’s really helped me to implement a lot of workflow tooling and optimization, but I also think I consult it more than I should.
    • Trying to ween myself off and focus on my manual work and building foundations, rather than just building more optimizations (maybe I’ll expand in a future post).