reaper@re4perz3d:~/writeups$ cat arcane-door.md
CyberSpark CTF — ArcaneDoor
Challenge Description
This challenge is part of a beginner-level Cyber Threat Intelligence (CTI) CTF. It focuses on analyzing a real-world APT campaign named ArcaneDoor, leveraging the MITRE ATT&CK framework, CVE databases, and threat actor profiling. Participants are expected to identify key TTPs (Tactics, Techniques, and Procedures), malicious software, and exploited vulnerabilities.
Questions & Answers
What is the name of the group of the threat actor attributed to the ArcaneDoor campaign?
Answer: UAT4356, STORM-1849
When was the ArcaneDoor campaign first observed? (Month year)
Answer: July 2023
What kind of devices were primarily targeted by ArcaneDoor? (2 words)
Answer: networking devices
What is the MITRE technique ID for exploitation of public-facing applications used in this campaign? (T****)
Answer: T1190
ArcaneDoor leveraged VPN servers under their control for C2. What is the relevant sub-technique ID for this tactic? (T****.***)
Answer: T1583.003
What are the ID and name of one of the two malicious software used in the ArcaneDoor campaign?
Answer: S1186–Line Dancer, S1188–Line Runner
What vulnerability (CVE) did the threat actors exploit to trigger the execution of Line Runner?
Answer: CVE-2024-20353
Which defensive mechanism did ArcaneDoor actors modify to bypass logging?
Answer: Authentication, Authorization, and Accounting (AAA) function
Skills Learned
- 🔍 Threat Actor Attribution: Learned how to trace and confirm threat actor identities using public threat reports and naming conventions (e.g., UAT/STORM).
- 🧠 MITRE ATT&CK Navigation: Gained fluency in identifying and interpreting techniques and sub-techniques like
T1190andT1583.003. - 🛡️ Vulnerability Research: Understood the process of mapping CVEs to threat campaigns using security advisories and databases.
- 🐾 TTP Analysis: Strengthened knowledge of Tactics, Techniques, and Procedures (TTPs) used by advanced actors in real-world scenarios.
- ⚙️ Defensive Evasion Techniques: Identified how AAA mechanisms can be tampered with to evade detection and logging.
Flag
Spark{W3lc0me_To_Thre4t_Int3ll1gence}