AGENT SKILL

performing-cloud-forensics-investigation

Collect and analyze cloud forensic evidence using AWS CLI, Azure CLI, or gcloud to snapshot volumes, capture instance metadata and security group configurations, and preserve cloud-native logs (CloudTrail, Activity Log, Audit Log). Use when investigating a suspected breach in AWS, Azure, or GCP, tracing unauthorized access through API logs, or analyzing a compromised VM, container, or serverless function.

License 许可
Apache-2.0
最近更新
2026-08-25
Stars
★ 0
Forks
0

01 前言 Front Matter

name
performing-cloud-forensics-investigation
tags
[ "forensics", "cloud-forensics", "aws", "azure", "gcp", "incident-response", "log-analysis" ]
author
mahipal
domain
cybersecurity
license
Apache-2.0
version
1.0
nist_csf
[ "RS.AN-03", "DE.AE-02", "RS.MA-01" ]
subdomain
digital-forensics
description
Collect and analyze cloud forensic evidence using AWS CLI, Azure CLI, or gcloud to snapshot volumes, capture instance metadata and security group configurations, and preserve cloud-native logs (CloudTrail, Activity Log, Audit Log). Use when investigating a suspected breach in AWS, Azure, or GCP, tracing unauthorized access through API logs, or analyzing a compromised VM, container, or serverless function.
mitre_attack
[ "T1005", "T1074", "T1119", "T1070", "T1078.004" ]

02 资源文件 3

03 SKILL.md 正文

Performing Cloud Forensics Investigation

When to Use

  • When investigating a security breach in AWS, Azure, or GCP cloud environments
  • For collecting volatile and non-volatile evidence from cloud infrastructure
  • When tracing unauthorized access through cloud service API logs
  • During incident response requiring preservation of cloud-based evidence
  • For analyzing compromised virtual machines, containers, or serverless functions

Prerequisites

  • Administrative access to the cloud account under investigation
  • AWS CLI, Azure CLI, or gcloud CLI configured with appropriate permissions
  • Understanding of cloud-native logging (CloudTrail, Activity Log, Audit Log)
  • Forensic workstation with cloud SDKs installed
  • Knowledge of IAM, networking, and compute services in target cloud
  • Evidence preservation procedures for cloud environments

Workflow

Step 1: Preserve Cloud Evidence and Establish Scope

# === AWS Evidence Preservation ===
# Snapshot compromised EC2 instance volumes
INSTANCE_ID="i-0abc123def456789"
VOLUME_IDS=$(aws ec2 describe-instances --instance-ids $INSTANCE_ID \
   --query 'Reservations[].Instances[].BlockDeviceMappings[].Ebs.VolumeId' --output text)

for vol in $VOLUME_IDS; do
   aws ec2 create-snapshot --volume-id $vol \
      --description "Forensic snapshot - Case 2024-001 - $(date -u)" \
      --tag-specifications "ResourceType=snapshot,Tags=[{Key=Case,Value=2024-001},{Key=Evidence,Value=true}]"
done

# Capture instance metadata
aws ec2 describe-instances --instance-ids $INSTANCE_ID \
   > /cases/case-2024-001/cloud/instance_metadata.json

# Capture security group rules
aws ec2 describe-security-groups --group-ids $(aws ec2 describe-instances \
   --instance-ids $INSTANCE_ID --query 'Reservations[].Instances[].SecurityGroups[].GroupId' --output text) \
   > /cases/case-2024-001/cloud/security_groups.json

# Capture network interfaces
aws ec2 describe-network-interfaces --filters "Name=attachment.instance-id,Values=$INSTANCE_ID" \
   > /cases/case-2024-001/cloud/network_interfaces.json

# Isolate the instance (replace security group with forensic isolation SG)
aws ec2 modify-instance-attribute --instance-id $INSTANCE_ID \
   --groups sg-forensic-isolation

# === Azure Evidence Preservation ===
# Snapshot a compromised VM disk
az snapshot create --resource-group forensics-rg \
   --name "case-2024-001-osdisk-snapshot" \
   --source "/subscriptions/SUB_ID/resourceGroups/RG/providers/Microsoft.Compute/disks/vm-osdisk"

# === GCP Evidence Preservation ===
gcloud compute disks snapshot compromised-disk \
   --snapshot-names="case-2024-001-forensic" \
   --zone=us-central1-a

Step 2: Collect Cloud API and Access Logs

# === AWS CloudTrail Logs ===
# Download CloudTrail events for the investigation period
aws cloudtrail lookup-events \
   --start-time "2024-01-15T00:00:00Z" \
   --end-time "2024-01-20T23:59:59Z" \
   --max-results 1000 \
   > /cases/case-2024-001/cloud/cloudtrail_events.json

# Filter for specific user activity
aws cloudtrail lookup-events \
   --lookup-attributes AttributeKey=Username,AttributeValue=compromised-user \
   --start-time "2024-01-15T00:00:00Z" \
   > /cases/case-2024-001/cloud/user_activity.json

# Download S3 access logs
aws s3 sync s3://my-cloudtrail-bucket/AWSLogs/ /cases/case-2024-001/cloud/cloudtrail_s3/

# Query CloudTrail with Athena for large-scale analysis
aws athena start-query-execution \
   --query-string "SELECT eventTime, eventName, userIdentity.arn, sourceIPAddress, errorCode
                   FROM cloudtrail_logs
                   WHERE eventTime BETWEEN '2024-01-15' AND '2024-01-20'
                   AND sourceIPAddress NOT IN ('10.0.0.0/8')
                   ORDER BY eventTime" \
   --result-configuration OutputLocation=s3://forensics-bucket/athena-results/

# === AWS VPC Flow Logs ===
aws logs filter-log-events \
   --log-group-name "vpc-flow-logs" \
   --start-time $(date -d "2024-01-15" +%s000) \
   --end-time $(date -d "2024-01-20" +%s000) \
   --filter-pattern "ACCEPT" \
   > /cases/case-2024-001/cloud/vpc_flow_logs.json

# === Azure Activity Log ===
az monitor activity-log list \
   --start-time "2024-01-15T00:00:00Z" \
   --end-time "2024-01-20T23:59:59Z" \
   --output json > /cases/case-2024-001/cloud/azure_activity.json

# === GCP Audit Logs ===
gcloud logging read 'logName="projects/PROJECT_ID/logs/cloudaudit.googleapis.com%2Factivity"
   AND timestamp>="2024-01-15T00:00:00Z"
   AND timestamp<="2024-01-20T23:59:59Z"' \
   --format=json > /cases/case-2024-001/cloud/gcp_audit.json

Step 3: Analyze IAM and Access Patterns

# Analyze compromised credentials usage
python3 << 'PYEOF'
import json
from collections import defaultdict

with open('/cases/case-2024-001/cloud/cloudtrail_events.json') as f:
    data = json.load(f)

# Analyze by source IP
ip_events = defaultdict(list)
error_events = []
critical_actions = []

for event in data.get('Events', []):
    ct = json.loads(event.get('CloudTrailEvent', '{}'))
    source_ip = ct.get('sourceIPAddress', 'Unknown')
    event_name = ct.get('eventName', 'Unknown')
    user_arn = ct.get('userIdentity', {}).get('arn', 'Unknown')
    error = ct.get('errorCode')
    timestamp = ct.get('eventTime', '')

    ip_events[source_ip].append(event_name)

    if error:
        error_events.append({'time': timestamp, 'action': event_name, 'error': error, 'ip': source_ip})

    # Flag critical actions
    critical = ['CreateUser', 'CreateAccessKey', 'AttachUserPolicy', 'CreateRole',
                'PutBucketPolicy', 'StopLogging', 'DeleteTrail', 'CreateKeyPair',
                'RunInstances', 'AuthorizeSecurityGroupIngress']
    if event_name in critical:
        critical_actions.append({'time': timestamp, 'action': event_name, 'user': user_arn, 'ip': source_ip})

print("=== SOURCE IP ANALYSIS ===")
for ip, events in sorted(ip_events.items(), key=lambda x: len(x[1]), reverse=True):
    print(f"  {ip}: {len(events)} events ({len(set(events))} unique actions)")

print(f"\n=== ACCESS ERRORS ({len(error_events)} total) ===")
for e in error_events[:10]:
    print(f"  [{e['time']}] {e['action']} -> {e['error']} from {e['ip']}")

print(f"\n=== CRITICAL ACTIONS ({len(critical_actions)} total) ===")
for a in critical_actions:
    print(f"  [{a['time']}] {a['action']} by {a['user']} from {a['ip']}")
PYEOF

Step 4: Acquire and Analyze VM Disk Image

# Create a forensic analysis instance from the snapshot
SNAPSHOT_ID="snap-0abc123def456789"

# Create volume from snapshot in isolated forensic VPC
FORENSIC_VOL=$(aws ec2 create-volume --snapshot-id $SNAPSHOT_ID \
   --availability-zone us-east-1a \
   --tag-specifications "ResourceType=volume,Tags=[{Key=Case,Value=2024-001}]" \
   --query 'VolumeId' --output text)

# Attach to forensic analysis instance (read-only mount)
aws ec2 attach-volume --volume-id $FORENSIC_VOL \
   --instance-id i-forensic-workstation \
   --device /dev/xvdf

# On the forensic instance, mount read-only
sudo mount -o ro /dev/xvdf1 /mnt/evidence

# Perform standard disk forensics on the mounted volume
# Extract logs, analyze file system, check for persistence
ls /mnt/evidence/var/log/
cp -r /mnt/evidence/var/log/ /cases/case-2024-001/cloud/vm_logs/
cp -r /mnt/evidence/etc/crontab /cases/case-2024-001/cloud/persistence/
cp -r /mnt/evidence/home/*/.ssh/ /cases/case-2024-001/cloud/ssh_keys/
cp -r /mnt/evidence/home/*/.bash_history /cases/case-2024-001/cloud/bash_history/

Step 5: Generate Cloud Forensics Report

# Compile findings into structured report
python3 << 'PYEOF'
report = """
CLOUD FORENSICS INVESTIGATION REPORT
======================================
Case: 2024-001
Cloud Provider: AWS (Account: 123456789012)
Region: us-east-1
Investigation Period: 2024-01-15 to 2024-01-20

EVIDENCE PRESERVED:
- EC2 Instance Snapshot: snap-0abc123def456789 (i-0abc123def456789)
- CloudTrail Logs: 2024-01-15 to 2024-01-20
- VPC Flow Logs: 2024-01-15 to 2024-01-20
- Instance Metadata: captured and hashed
- Security Group Configuration: captured at time of isolation

FINDINGS:
1. Initial Access:
   - Compromised IAM access key AKIA... used from IP 203.0.113.45
   - First unauthorized API call: 2024-01-15 14:32:00 UTC
   - IP geolocation: Foreign jurisdiction (not company IP range)

2. Persistence:
   - New IAM user 'backup-admin' created with AdministratorAccess
   - New access key pair generated for backup-admin
   - SSH key added to EC2 instance authorized_keys

3. Lateral Movement:
   - S3 bucket policies modified to allow public access
   - Security group rules modified to allow SSH from 0.0.0.0/0
   - 3 additional EC2 instances launched for crypto-mining

4. Data Exfiltration:
   - S3 bucket 'company-confidential' accessed 234 times
   - 12 GB of data downloaded via GetObject API calls
   - Data transferred to external IP 185.x.x.x

5. Anti-Forensics:
   - CloudTrail logging disabled at 2024-01-18 03:00 UTC
   - CloudWatch log groups deleted

RECOMMENDATIONS:
- Rotate all IAM credentials immediately
- Enable MFA on all accounts
- Restore CloudTrail logging
- Review and restrict S3 bucket policies
- Implement GuardDuty for continuous monitoring
"""

with open('/cases/case-2024-001/cloud/cloud_forensics_report.txt', 'w') as f:
    f.write(report)
print(report)
PYEOF

Key Concepts

Concept Description
Cloud API logging Service logs recording all API calls (CloudTrail, Activity Log, Audit Log)
Volume snapshots Point-in-time copies of cloud disk volumes for forensic preservation
VPC Flow Logs Network traffic metadata logs showing source, destination, and action
IAM credential compromise Unauthorized use of access keys, tokens, or assumed roles
Instance metadata EC2/VM configuration data including network, storage, and security settings
Shared responsibility Cloud provider secures infrastructure; customer secures data and access
Evidence volatility Cloud resources can be terminated; evidence must be preserved quickly
Multi-region artifacts Attacks may span regions requiring cross-region log collection

Tools & Systems

Tool Purpose
AWS CLI Command-line interface for AWS service interaction and log collection
CloudTrail AWS API call logging service for investigation and auditing
Azure Monitor Azure logging and diagnostics platform
GCP Cloud Logging Google Cloud audit and access logging service
Athena AWS serverless SQL query service for analyzing CloudTrail logs at scale
Prowler Open-source AWS security assessment and forensic collection tool
ScoutSuite Multi-cloud security auditing tool
CADO Response Cloud-native digital forensics and incident response platform

Common Scenarios

Scenario 1: Compromised IAM Access Keys Identify the compromised key in CloudTrail, trace all API calls made with the key, determine the source IPs and actions taken, check for persistence mechanisms (new users, roles, keys), revoke the compromised credentials, assess data access scope.

Scenario 2: Cryptojacking on EC2 Instances Detect unauthorized instance launches in CloudTrail, snapshot the mining instances for analysis, examine security group changes that allowed C2 communication, identify the initial access vector (stolen keys, SSRF), calculate resource costs incurred.

Scenario 3: S3 Data Breach Analyze S3 access logs and CloudTrail for GetObject/PutBucketPolicy events, identify who modified bucket policies to allow public access, determine the scope of data exposure, check for data downloads from unauthorized IPs, assess regulatory reporting requirements.

Scenario 4: Container Escape in EKS/AKS/GKE Collect Kubernetes audit logs and cloud provider logs, analyze pod creation events for privilege escalation attempts, examine node-level logs for container escape evidence, check for unauthorized access to cloud metadata service (169.254.169.254), trace lateral movement to cloud APIs.

Output Format

Cloud Forensics Summary:
  Cloud: AWS (us-east-1) Account: 123456789012
  Investigation: 2024-01-15 to 2024-01-20
  Incident Type: IAM Credential Compromise + Data Exfiltration

  Evidence Collected:
    EBS Snapshots:    3 volumes preserved
    CloudTrail Events: 12,456 (1,234 from attacker IP)
    VPC Flow Logs:    45,678 records
    S3 Access Logs:   2,345 entries

  Attack Timeline:
    2024-01-15 14:32 - Compromised access key first used from 203.0.113.45
    2024-01-15 14:45 - New IAM user created with admin privileges
    2024-01-16 02:00 - S3 bucket policy modified (public access enabled)
    2024-01-16 03:00 - 12 GB downloaded from company-confidential bucket
    2024-01-18 03:00 - CloudTrail logging disabled

  Impact Assessment:
    Data Exposed: 12 GB from 3 S3 buckets
    Resources Created: 3 EC2 instances (crypto mining)
    Estimated Cost: $4,500 in unauthorized compute

04 同仓库更多 Skills

vikasvardhanv
vikasvardhanv/jerry_skills

syllabus

Generates a curated supplementary reading list from any course syllabus using Consensus academic search. Grill-me intake (syllabus input format + course audience + year range) plus a grouping forcing-options checkpoint before any search runs — so the reading list matches the course's level and recency need. Parses the syllabus to extract topics and learning outcomes, searches Consensus for recent peer-reviewed papers per topic, and produces a professionally formatted .docx with clickable Consensus links, plain-language summaries calibrated to audience level, and Bloom-higher-order discussion questions tied to course learning goals. Use when the user uploads a syllabus, course outline, or curriculum document and wants supplementary readings (e.g., 'create a reading list from this syllabus', 'find recent papers for my course') — even casual mentions with a syllabus attached should trigger this skill.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

pulse

Multi-source recency research skill that takes the pulse of any topic across Reddit, Hacker News, the open web, and optionally X/Twitter within a configurable recent window (default 30 days). Forcing intake clarifies topic specificity, angle (trend/sentiment/problems/opportunities/comparison), time window, and platform scope before searching. Returns a synthesized briefing with citations, engagement metrics, and cross-platform pattern analysis. Use when the user requests multi-source recency intelligence on a topic (e.g., 'pulse on [topic]', 'what's happening with [topic]', 'what are people saying about [topic]', 'current conversation about [topic]', 'take the pulse of [topic]', 'trending: [topic]', 'find me info on [topic]'), and for competitor research, trend discovery, tool comparisons, and audience sentiment analysis.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

patent

Patent prior-art and landscape intelligence skill — not generic patent help. Commits to one of five sub-use-cases via forcing intake (novelty search / freedom-to-operate / competitive landscape / acquisition diligence / litigation prior-art) before any search runs. Searches Google Patents, Espacenet, USPTO, and optionally Lens.org for citation-graph signals. Output is an editable Word document (.docx) with verdict, ranked closest art (claim-text extracted), CPC-class-aware landscape, family-resolved hits, geographic coverage, FTO flags where applicable, strategy recommendations, and full audit log. Use when the user asks for patent searching or analysis (e.g., 'prior art search for [invention]', 'freedom to operate analysis for [product]'). Produces search signal, not legal advice — always recommends consulting a patent attorney before filing or licensing decisions. Trademark, copyright, and trade-secret questions are out of scope.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

notebooklm

Browser automation skill for controlling Google's NotebookLM. Use when the user wants anything done in NotebookLM (e.g., 'open NotebookLM', 'check my [name] notebook', 'ask my notebook about X', 'add [source] to NotebookLM', 'generate a Video Overview from my notebook', 'use NotebookLM Studio'). Handles reading and querying notebooks, adding sources (URLs, text, files, YouTube links, synthesized content), generating Studio outputs (Audio/Video Overviews, Mind Maps, Reports incl. Briefing Doc/Study Guide/FAQ, Flashcards, Quiz, slide decks, infographics — discover the exact set from the live Studio panel; the UI evolves fast), and creating new notebooks. Requires browser automation environment — fails gracefully when unavailable.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

litreview

Academic literature orientation skill that searches papers via free keyless APIs (PubMed E-utilities + OpenAlex) by default — with the Consensus MCP as an optional enhancement lane when connected — builds a strategic search plan using PICO (default) or SPIDER / Decomposition / hybrid as fallbacks, and synthesizes findings into a formatted Word (.docx) research guide. Grill-me intake (research question specificity + framework hint + tentative depth) before the recon search; a second forcing checkpoint after Phase 2 confirms framework + sub-areas + depth before searches consume budget. Configurable depth (5/10/20 queries) controls coverage vs. speed. Output is a 'launching pad' — an orientation guide that lets a researcher dive in confidently, not a finished review. Use when the user starts literature-oriented research (e.g., 'litreview on [topic]', 'literature review on [topic]', 'I'm starting a literature review on X', 'I'm writing a paper on X', 'help me research X', 'I'm doing research on X', 'can you help me research X'). Do NOT use for single one-off paper searches wanting a quick list — that's a plain PubMed/OpenAlex (or Consensus) query.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

grants

NIH grant research skill for clinical researchers. Grill-me intake (research idea + career stage + preliminary data + environment + submission posture + known institute targets) locks down the funding strategy before any search runs. Runs a 5-facet Consensus positioning analysis (with draft Significance/Innovation language), maps the research to the right NIH institutes and study sections via RePORTER, finds NOSIs and funded overlap, and produces an editable Word document (.docx) with budget/scope-aware mechanism recommendations, submission timelines, and a mandatory program officer recommendation. Use when the user asks about research funding or makes any grant-related request (e.g., 'grants for [topic]', 'find grants for my research idea', 'what grants match my research', 'help me find NIH funding', 'grant opportunities for my research'). NIH-only scope — non-NIH funders (PCORI, DOD CDMRP, VA, foundations) are out of scope and flagged at intake.

★ 0