Arcana IR documentation · offline copy Nederlands

RB-TRIAGE-001: Email Triage & Header Analysis

Document Control

Attribute Value Date
Document Name Email Triage & Header Analysis
Version [Enter version number] [Enter date]
Owner [Enter owner/team] [Enter date]
Status [Draft/Approved/Retired] [Enter date]
Next Review Date [Enter next review date] [Enter date]
Approvals [Enter approver(s)] [Enter date]
Change Summary Initial creation [Enter date]

1. Prerequisites

Required Tools:

2. Step-by-Step Instructions

  1. Collect Full Email Data
    • Retrieve full email headers (not partial)
    • Capture:
      • Sender (From, Reply-To)
      • Subject
      • Timestamp
      • Message-ID
  2. Analyse Authentication Results - Check:
    • SPF → pass/fail
    • DKIM → pass/fail
    • DMARC → pass/fail/alignment.
    • Indicators:
      • SPF fail or softfail
      • DKIM missing or invalid
      • DMARC fail
  3. Validate Sender Identity
    • Compare:
      • Display Name vs Actual Email Address
      • Domain similarity (typosquatting, homoglyphs)
    • Check:
      • Newly registered domains
      • External sender spoofing internal user
  4. Review Email Routing
    • Analyse Received: headers:
      • Originating IP
      • Sending infrastructure
      • Unusual relays or geolocation anomalies
  5. Extract Indicators
    • Pull out:
    • All URLs (including obfuscated URLs) Attachment names + hashes
    • Reply-To domain (often different)
  6. Perform Initial Reputation Checks
    • URLS → VirusTotal/URLScan
    • Domains → passive DNS / reputation
    • File hashes → malware databases
  7. Assess Email Content:
    • Look for:
      • urgency/pressure tactics
      • credential prompts
      • financial requests
      • suspicious attachments
      • link text vs actual destination mismatch

3. Post-Action

Contributor

Vishal Thakur GitHub: https://github.com/malienist

Jayden Vo GitHub: https://github.com/jayden-vo

Contributed to the Arcana Incident Response Documentation Framework.