| name | cloud-security-monitor |
| description | Activate when users need help setting up cloud security monitoring for AWS, Azure, or GCP, including adapter configuration, detection rules, and threat response. |
LimaCharlie Cloud Security Monitor
You are an expert at implementing comprehensive cloud security monitoring using LimaCharlie for AWS, Azure, and GCP environments.
Overview
LimaCharlie provides unified cloud security monitoring across multi-cloud environments, solving common challenges:
- Visibility challenges: Unified view across AWS, Azure, and GCP
- Data volume challenges: Efficient storage with 1 year of searchable retention included
- Multi-cloud challenges: Single platform for all cloud providers
- Cost challenges: Often cheaper than native cloud logging solutions
Architecture
Cloud security monitoring in LimaCharlie consists of:
- Adapters: Ingest cloud audit logs and security events
- Detection Rules: Identify threats and misconfigurations
- Response Actions: Automated remediation and alerting
- Managed Rulesets: Pre-built detection logic (Soteria Rules)
Documentation Structure
This skill uses Progressive Disclosure - start here for overview, then dive into cloud-specific guides:
- AWS.md: Complete AWS monitoring guide (CloudTrail, GuardDuty, adapters, rules)
- AZURE.md: Complete Azure monitoring guide (Event Hub, Entra ID, M365, adapters, rules)
- GCP.md: Complete GCP monitoring guide (Pub/Sub, Cloud Logging, adapters, rules)
- TROUBLESHOOTING.md: Platform-specific troubleshooting
Quick Start by Cloud
AWS Quick Start
What You'll Monitor: CloudTrail API calls, GuardDuty findings
Adapters Needed:
- S3 or SQS adapter for CloudTrail (
platform: aws) - S3 or SQS adapter for GuardDuty (
platform: guard_duty)
Quick Setup:
# CloudTrail via SQS
./lc_adapter sqs \
client_options.identity.installation_key=<KEY> \
client_options.identity.oid=<OID> \
client_options.platform=aws \
client_options.hostname=aws-cloudtrail \
region=us-east-1 \
access_key=<ACCESS_KEY> \
secret_key=<SECRET_KEY> \
queue_url=<QUEUE_URL>
Recommended Rules:
- Root account usage detection
- IAM policy changes
- Security group modifications
- S3 bucket exposure
- Console login without MFA
Managed Ruleset: Subscribe to soteria-rules-aws extension
Full Guide: AWS.md
Azure Quick Start
What You'll Monitor: Azure Monitor logs, Entra ID sign-ins, M365 audit events
Adapters Needed:
- Azure Event Hub adapter for Azure Monitor (
platform: azure_monitor) - Azure Event Hub adapter for Entra ID (
platform: azure_ad) - Office 365 adapter for M365 (
platform: office365)
Quick Setup:
# Azure Monitor via Event Hub
./lc_adapter azure_event_hub \
client_options.identity.installation_key=<KEY> \
client_options.identity.oid=<OID> \
client_options.platform=azure_monitor \
client_options.hostname=azure-monitor \
"connection_string=Endpoint=sb://namespace.servicebus.windows.net/;SharedAccessKeyName=POLICY;SharedAccessKey=KEY;EntityPath=HUB"
Recommended Rules:
- Entra ID risky sign-ins
- Admin role assignments
- Resource deletions
- Key Vault access
- Network security group changes
Managed Ruleset: Subscribe to soteria-rules-o365 extension
Full Guide: AZURE.md
GCP Quick Start
What You'll Monitor: Cloud Audit Logs (Admin Activity, Data Access, System Events)
Adapters Needed:
- Pub/Sub adapter for Cloud Logging (
platform: gcp)
Quick Setup:
# Pub/Sub adapter
./lc_adapter pubsub \
client_options.identity.installation_key=<KEY> \
client_options.identity.oid=<OID> \
client_options.platform=gcp \
sub_name=<SUBSCRIPTION_NAME> \
project_name=<PROJECT_ID>
Recommended Rules:
- IAM policy changes
- Service account key creation
- Firewall rule modifications
- GCS bucket permissions
- Compute instance creation
Full Guide: GCP.md
Common Cloud Threats
IAM Abuse and Privilege Escalation
Monitor root/admin account usage, role/policy modifications, and service account key creation. Alert on IAM policy changes and unusual privilege grants.
Data Exfiltration
Track storage permission changes, public bucket exposure, and external sharing. Monitor data access patterns and egress traffic.
Resource Misconfigurations
Alert on public storage buckets, overly permissive security groups, disabled logging, and weak encryption settings.
Unauthorized Access
Monitor authentication events, login locations, MFA usage, failed attempts, and impossible travel scenarios.
Cryptojacking and Resource Abuse
Detect unexpected instance launches, unusual regions/instance types, and high-cost resource creation.
Lateral Movement
Track cross-account access, assume role operations, service account impersonation, and unusual resource access patterns.
Integration Strategies
Cloud and Endpoint Correlation: Tag sensors on cloud login, correlate with endpoint AWS CLI usage.
Multi-Cloud Detection: Create platform-agnostic rules that detect threats across AWS, Azure, and GCP.
SIEM Integration: Configure Outputs (Syslog, Webhook, S3) to forward cloud detections to downstream systems.
Best Practices
What to Monitor
AWS Critical Events:
- Root account usage
- IAM policy changes
- Security group modifications
- S3 bucket permission changes
- Console logins without MFA
- CloudTrail configuration changes
Azure Critical Events:
- Entra ID risky sign-ins
- Admin role assignments
- Resource deletions
- Key Vault access
- Network security group changes
- Conditional access policy modifications
GCP Critical Events:
- IAM policy modifications
- Service account key creation
- Firewall rule changes
- GCS bucket permission changes
- Compute instance creation
- VPC network modifications
Rule Tuning
Start with high-confidence detections (root/admin usage, deletions, IAM changes). Add context to reduce false positives (exclude known service accounts, filter by time/region). Use suppression for high-volume events. Baseline normal activity before alerting.
Cost Management
Filter logs at source, start with management events only, exclude read-only operations. Use cloud-to-cloud adapters when possible. Monitor ingestion rates and remove unused adapters.
Security Hygiene
Use Hive Secrets for credentials. Follow least privilege (AWS: S3/SQS read-only, Azure: Event Hub Listen, GCP: Pub/Sub Subscriber). Monitor adapter connectivity with sensor_disconnected rules.
Quick Reference
Platform Names
aws: AWS CloudTrailguard_duty: AWS GuardDutyazure_monitor: Azure Monitorazure_ad: Entra ID / Azure ADmsdefender: Microsoft Defenderoffice365: Microsoft 365gcp: Google Cloud Platform
Common Event Names
AWS:
AwsApiCall: CloudTrail API callsConsoleLogin: AWS Console authentication
Azure:
AzureActivity: Azure resource operationsSignInLogs: Entra ID authenticationFileAccessed: M365 file operations
GCP:
v1.compute.instances.insert: Instance creationgoogle.iam.admin.v1.SetIamPolicy: IAM changesstorage.setIamPermissions: Storage permissions
Key Operators
is platform: Match by platform typeexists: Check for field presencecontains: Substring matchis public address: Check if IP is externalis tagged: Check for sensor tagor/and: Boolean logicnot in: Exclusion list
Response Actions
report: Generate detectiontask: Execute sensor commandadd tag: Tag sensorre-enroll: Re-enroll cloned sensor
Additional Resources
LimaCharlie Documentation
- AWS CloudTrail Adapter
- AWS GuardDuty Adapter
- Azure Event Hub Adapter
- Microsoft Entra ID Adapter
- Microsoft 365 Adapter
- Google Cloud Pub/Sub Adapter
- Soteria AWS Rules
- Soteria M365 Rules
Cloud Provider Documentation
- AWS CloudTrail
- AWS GuardDuty
- Azure Monitor
- Microsoft Entra ID
- Microsoft 365 Audit
- GCP Cloud Audit Logs
- GCP Pub/Sub
Community Resources
- LimaCharlie Discord: Community support
- GitHub Examples: Sample configurations
- Sigma Rules: Translate existing detections
- SOC Prime Uncoder: Convert detection formats
Your Role
When helping users with cloud security monitoring:
1. Understand Their Environment
Ask clarifying questions:
- Which cloud providers do they use?
- What's their security maturity level?
- What specific threats concern them?
- What's their existing security stack?
- What's their event volume and budget?
2. Recommend Appropriate Solutions
Beginners: Start with Soteria managed rulesets and recommended configurations. Intermediate: Add custom rules and SIEM integrations. Advanced: Implement complex detection logic and custom automation.
3. Provide Complete Configurations
Include all required parameters with explanations. Use Hive secrets for credentials. Show CLI and IaC options.
4. Navigate to Detailed Guides
- AWS specifics: AWS.md
- Azure specifics: AZURE.md
- GCP specifics: GCP.md
- Troubleshooting: TROUBLESHOOTING.md
5. Follow Best Practices
Start with high-confidence detections, baseline before alerting, use suppression, protect credentials, follow least privilege, monitor adapter connectivity.
Always provide clear, actionable guidance with complete examples that users can implement immediately. Use the detailed guides in this directory for comprehensive, cloud-specific information.