Skip to main content
The OCI Cloud Guard MCP server provides tools for interacting with Oracle Cloud Infrastructure Cloud Guard service, enabling management of security problems and threat detection.

Installation

uvx oracle.oci-cloud-guard-mcp-server

Running the Server

STDIO Transport Mode

uvx oracle.oci-cloud-guard-mcp-server

HTTP Streaming Transport Mode

ORACLE_MCP_HOST=<hostname/IP address> ORACLE_MCP_PORT=<port number> uvx oracle.oci-cloud-guard-mcp-server

Available Tools

The server provides the following tools for security management:
Tool NameDescription
list_problemsList security problems in a given compartment
get_problem_detailsGet detailed information about a problem using its OCID
update_problem_statusUpdate the status of a security problem

Usage Examples

List Security Problems

List all security problems in my compartment
Retrieves all detected security problems with their risk levels, states, and descriptions.
Show me all critical security problems
Filters problems by risk level to focus on highest priority issues.
List open security problems from the last 7 days
Filters by time range and status to track recent detections.

Get Problem Details

Show me details for problem ocid1.cloudguardproblem.oc1.phx.example
Retrieves comprehensive information about a specific security problem including:
  • Risk level and score
  • Problem description
  • Affected resources
  • Detection time
  • Recommendations
  • Related responder activities

Update Problem Status

Mark problem ocid1.cloudguardproblem.oc1.phx.example as resolved
Updates problem status after remediation or investigation.
Dismiss problem ocid1.cloudguardproblem.oc1.phx.example as false positive
Allows dismissing problems that are not actual security issues.

Understanding OCI Cloud Guard

What is Cloud Guard?

Cloud Guard is Oracle’s cloud-native security monitoring and threat detection service that:
  • Detects security weaknesses and threats
  • Monitors configuration compliance
  • Responds automatically to security problems
  • Reports on security posture
  • Prevents risky configurations

Key Concepts

Problems Security issues detected by Cloud Guard:
  • Misconfigurations violating security best practices
  • Suspicious or anomalous activities
  • Policy violations
  • Risk-scored and prioritized
Detectors Rules that identify security problems:
  • Configuration detectors - Check resource configurations
  • Activity detectors - Monitor user and system activities
  • Oracle-managed detector recipes
  • Custom detector configurations
Responders Automated remediation actions:
  • Respond to detected problems automatically
  • Predefined remediation actions
  • Disable resources, modify configurations
  • Can be enabled/disabled per responder rule
Targets Compartments monitored by Cloud Guard:
  • Define scope of monitoring
  • Assign detector and responder recipes
  • Hierarchical - child compartments inherit settings

Risk Levels

Critical
  • Severe security risks requiring immediate attention
  • Examples: Public buckets with sensitive data, open databases
High
  • Significant security concerns
  • Examples: Overly permissive IAM policies, disabled security features
Medium
  • Notable security issues
  • Examples: Missing encryption, non-compliant configurations
Low
  • Minor security findings
  • Examples: Informational alerts, best practice recommendations
Minor
  • Minimal risk items
  • Examples: Optimization suggestions

Problem Lifecycle

  1. Open - Newly detected problem
  2. In Progress - Under investigation or remediation
  3. Resolved - Issue has been fixed
  4. Dismissed - Marked as false positive or accepted risk
  5. Deleted - Removed from Cloud Guard

Common Security Problems

Identity & Access Management

  • Users with overly broad permissions
  • Service accounts with admin access
  • Long-lived API keys
  • Missing MFA on privileged accounts
  • Inactive users with active credentials

Network Security

  • Security lists allowing unrestricted access (0.0.0.0/0)
  • Public subnets with sensitive resources
  • Missing network security groups
  • Unencrypted traffic
  • Open management ports (SSH, RDP)

Data Protection

  • Public object storage buckets
  • Unencrypted volumes and databases
  • Object versioning disabled
  • Missing backup policies
  • Data not in approved regions

Compute Security

  • Instances with public IPs in private subnets
  • Security agents disabled
  • Non-compliant instance images
  • Missing OS patches
  • Exposed management interfaces

Database Security

  • Databases accessible from internet
  • Weak authentication settings
  • Audit logging disabled
  • Unencrypted data at rest
  • Missing database backups

Authentication

The server uses OCI CLI configuration from ~/.oci/config:
oci setup config

Required Permissions

Your OCI user or instance principal needs these IAM permissions: Read Problems:
Allow group CloudGuardAnalysts to read cloud-guard-problems in tenancy
Allow group CloudGuardAnalysts to read cloud-guard-targets in tenancy
Manage Problems:
Allow group CloudGuardOperators to manage cloud-guard-problems in tenancy
Allow group CloudGuardOperators to read cloud-guard-targets in tenancy
Full Cloud Guard Admin:
Allow group CloudGuardAdmins to manage cloud-guard-family in tenancy
Security NoticeAll actions are performed with the permissions of the configured OCI CLI profile. We advise:
  • Least-privilege IAM setup
  • Secure credential management
  • Safe network practices
  • Secure logging
  • Never expose secrets in logs or responses
  • Review security findings promptly

Common Use Cases

Security Monitoring

  • Monitor security posture across tenancy
  • Track new security problems
  • Prioritize remediation by risk level
  • Measure security metrics over time

Compliance Management

  • Detect configuration drift from baselines
  • Verify compliance with security standards
  • Generate compliance reports
  • Track remediation progress

Incident Response

  • Investigate security alerts
  • Identify affected resources
  • Track problem remediation
  • Document security incidents

Automated Remediation

  • Enable responders for auto-remediation
  • Monitor responder execution
  • Review automated changes
  • Adjust responder configurations

Security Operations

  • Daily security problem review
  • Triage new findings
  • Assign problems to teams
  • Track SLA compliance

Best Practices

Problem Management

  • Review new problems daily
  • Prioritize by risk level
  • Document remediation actions
  • Track mean time to resolution
  • Don’t dismiss problems without investigation

Detection Configuration

  • Enable Cloud Guard on all compartments
  • Use Oracle-managed detector recipes
  • Customize detectors for your requirements
  • Regularly review detector configurations
  • Test detector rules periodically

Responder Configuration

  • Start with responders in “Test” mode
  • Review automated actions before enabling
  • Configure notifications for responder actions
  • Monitor responder execution logs
  • Adjust based on operational impact

Integration

  • Route problems to SIEM systems
  • Integrate with ticketing systems
  • Set up notifications for critical problems
  • Create dashboards for security metrics
  • Automate reporting

Organizational

  • Define problem ownership
  • Establish SLAs for remediation
  • Create runbooks for common problems
  • Train teams on Cloud Guard
  • Regular security posture reviews

Remediation Strategies

Immediate Actions (Critical/High)

  1. Verify the problem is valid
  2. Assess impact and risk
  3. Implement immediate mitigation
  4. Fix root cause
  5. Verify resolution
  6. Update problem status

Planned Remediation (Medium/Low)

  1. Review and prioritize
  2. Schedule remediation work
  3. Plan changes and testing
  4. Implement fixes
  5. Validate and close problems

False Positives

  1. Investigate thoroughly
  2. Document why it’s not a real issue
  3. Consider detector tuning
  4. Dismiss with detailed comment
  5. Review dismissals periodically

Monitoring & Reporting

Key Metrics

  • Total open problems
  • Problems by risk level
  • Mean time to detection (MTTD)
  • Mean time to resolution (MTTR)
  • Problems opened vs. closed trends
  • Top problem types
  • Problems by compartment/resource

Dashboards

Create dashboards showing:
  • Security posture score
  • Problem trends over time
  • Risk distribution
  • Remediation velocity
  • Responder activity

Reports

Generate regular reports:
  • Weekly security summary
  • Monthly compliance reports
  • Quarterly security trends
  • Audit reports for compliance

Troubleshooting

Problem Not Appearing

Check:
  • Cloud Guard is enabled on target compartment
  • Detector rule is enabled
  • Resource is in monitored compartment
  • Sufficient time for detection (can take hours)
  • IAM permissions allow problem creation

Cannot Update Problem Status

Possible causes:
  • Insufficient permissions
  • Problem OCID is incorrect
  • Problem already in that status
  • Network connectivity issues
Solutions:
  • Verify IAM policies include manage cloud-guard-problems
  • Confirm problem OCID
  • Check current problem status
  • Test API connectivity

Too Many False Positives

Optimization:
  • Review and tune detector conditions
  • Create custom detector recipes
  • Exclude specific resources if appropriate
  • Adjust risk thresholds
  • Work with Oracle Support on detector rules

Responder Not Executing

Check:
  • Responder rule is enabled
  • IAM policies allow responder execution
  • Resource type supports responder action
  • Responder execution logs for errors
  • Resource state allows action

Integration Patterns

SIEM Integration

Cloud Guard Problems

Service Connector Hub

Streaming/Functions

SIEM System

Automated Ticketing

Cloud Guard Problem Detected

Events Rule Triggered

Function Executes

Ticket Created in ITSM

Notification Workflow

Critical Problem

Responder Executes

Notification Sent

Security Team Alerted

Additional Resources

Build docs developers (and LLMs) love