incident-response

Guides security incident response using NIST 800-61r3, forensics, and ATT&CK mapping.

1|Updated Jul 17, 2026
One-click install
npx skills add https://github.com/anonymous99-Rise/multi-CyberSecurity --skill incident-response-anonymous99-rise
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: incident-response
Source: https://github.com/anonymous99-Rise/multi-CyberSecurity/tree/main/Skills20260809/incident-response
Command: npx skills add https://github.com/anonymous99-Rise/multi-CyberSecurity --skill incident-response-anonymous99-rise

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When a security breach occurs, teams often lack a structured playbook for triaging alerts, preserving volatile evidence, reconstructing attack chains, and coordinating containment across hosts, cloud, and containers. This Skill provides a complete incident response methodology so responders act quickly and consistently instead of improvising under pressure. ## Core Features & Use Cases - Six-Phase Response Framework: Implements NIST SP 800-61r3 (preparation through post-incident) with severity grading (P0-P4), golden-hour timelines, and RACI coordination. - Deep Forensics Coverage: Provides concrete commands for Windows/Linux memory capture, Volatility 3 analysis, KAPE triage, fileless attack hunting, and evidence chain-of-custody handling. - Specialized Scenarios: Covers ransomware response, cloud (AWS/Azure/GCP) and Kubernetes incident handling, hypothesis-driven threat hunting, and LLM security incident response (prompt injection, tool poisoning). - Use Case: An EDR alert shows suspicious PowerShell execution on a database server at 2 AM. Use this Skill to grade the incident, capture memory and logs in the correct volatility order, map findings to MITRE ATT&CK, contain the host, and produce a post-incident report. ## Quick Start Ask the AI to walk you through responding to a confirmed intrusion on a Windows server, starting with evidence collection and containment steps.

Frequently Asked Questions about incident-response

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I respond to a security incident step by step?

Follow the NIST SP 800-61r3 six phases: preparation, detection and analysis, containment, eradication, recovery, and post-incident activity. Start by grading severity (P0-P4), collect volatile evidence like memory before isolating hosts, then rebuild the attack timeline and remove persistence mechanisms.

How to analyze a memory dump with Volatility 3?

Capture RAM first with winpmem, DumpIt, or LiME, then run Volatility 3 plugins such as windows.pslist, windows.pstree, windows.netscan, and windows.malfind to find hidden processes, injected code, and network connections. Use windows.cmdline and windows.hashdump for command lines and credentials.

What is the correct order for collecting forensic evidence?

Collect evidence in order of decreasing volatility: memory first, then network connections, running processes, disk artifacts, and logs. Never reboot, run antivirus scans, or delete files before acquisition, and compute SHA-256 hashes of every collected artifact for chain of custody.

How do I handle a ransomware incident response?

Isolate affected hosts within 30 minutes, preserve the ransom note and encrypted samples, and check no-more-ransom.org for decryptors before attempting recovery. Involve legal early since payment may violate sanctions, and assess data exfiltration scope because double extortion is now common.

Does incident response differ for Kubernetes and cloud environments?

Yes, containers are highly ephemeral so capture kubectl logs --previous, container filesystem diffs, and volume snapshots before pods are rebuilt. For cloud, collect both VM-level evidence and control-plane audit logs like CloudTrail, since attacker actions such as IAM changes are invisible from inside the VM.

What are the limitations of using LLMs for incident response?

LLMs can hallucinate log entries or evidence, so every AI-generated conclusion must include verifiable log references and human review. Sensitive data should be redacted before entering model context, and high-impact actions like host isolation or account disabling require human or policy-engine approval.