<< All versions

Skill v1.0.0

currentAutomated scan100/100
ahmadvh/octochains/threat-intel-triage
──Details
PublishedOctober 16, 2026 at 05:19 PM
Content Hashsha256:afef00ceb82dea2e...
Git SHA
──Files
Files (1 file, 1.2 KB)
SKILL.md1.2 KBactive
SKILL.md · 39 lines · 1.2 KB

name: threat-intel-triage description: Identifies indicators of compromise, anomalous network/endpoint behavior, and active intrusion patterns from the provided data. version: 1.0.0


Objective

Analyze security logs to identify suspicious activity and potential cyber attacks.

Execution Protocol

  1. Threat Detection: Identify unusual activity that may indicate a security attack, such as unusual outbound traffic, privilege escalation patterns, or off-hours access.
  2. Severity Prioritization: If several suspicious activities are found, prioritize the most critical findings based on their severity.
  3. Framework Mapping: If possible, relate the suspicious activity to the appropriate MITRE ATT&CK tactic.

Constraints

  • Report only findings supported by the provided evidence.
  • Mention if any important information is missing.
  • Don't assume an attack occurred if there is insufficient supporting evidence.

Output Format

Overall Threat Level: [LOW | MEDIUM | HIGH | CRITICAL]

Priority Findings:

  • [Finding 1]
  • [Finding 2]

Indicators of Compromise:

  • [Observed IoCs]

MITRE ATT&CK Mapping:

  • [Relevant tactic categories]

Recommended Next Steps:

  • [Immediate investigation or containment actions]
All versions