Skip to main content

Model Context Protocol (MCP)

Early Access

Try our newest feature! Only available through reach out. Contact us to join the early access program and get priority support.

Unizo offers an MCP server that integrates public cloud infrastructure platforms with any LLM provider supporting the MCP protocol. This enables your AI agent to perform resource management, analyze infrastructure patterns, and automate cloud operations across AWS, Azure, Google Cloud, and other providers via a single MCP server.

Supported Tools & Use Cases

The following tools are available in the Public Cloud (Infra) MCP Server:

Tool NameDescription
list_resourcesBrowse and search cloud resources across regions and providers
get_resource_detailsRetrieve comprehensive resource information including configuration and security settings
provision_resourceCreate new infrastructure resources with security best practices
modify_configurationUpdate resource configurations and apply security patches
analyze_securityAnalyze infrastructure security posture and identify vulnerabilities
monitor_healthMonitor resource health, performance, and security metrics

list_resources

Browse and search cloud resources across regions and providers

Parameters:

  • integration: Target cloud provider integration (required)
  • resource_type: Filter by resource type (vm, container, database, network)
  • region: Specific region or 'all' for global search
  • tags: Filter by resource tags
  • state: Filter by resource state (running, stopped, terminated)
  • page: Pagination control
  • limit: Number of results per page

get_resource_details

Retrieve comprehensive resource information including configuration and security settings

Parameters:

  • integration: Target cloud provider integration (required)
  • resource_id: Unique resource identifier (required)
  • include_metrics: Include performance and utilization metrics
  • include_configuration: Include detailed configuration settings
  • include_security_groups: Include network security rules
  • include_compliance: Include compliance status and findings

provision_resource

Create new infrastructure resources with security best practices

Parameters:

  • integration: Target cloud provider integration (required)
  • resource_type: Type of resource to create (required)
  • configuration: Resource configuration specifications (required)
  • region: Deployment region (required)
  • tags: Resource tags for organization and compliance
  • security_profile: Predefined security configuration template

modify_configuration

Update resource configurations and apply security patches

Parameters:

  • integration: Target cloud provider integration (required)
  • resource_id: Target resource identifier (required)
  • changes: Configuration changes to apply (required)
  • backup: Create backup before modification
  • validation: Validate changes before applying
  • rollback_on_failure: Automatic rollback if validation fails

analyze_security

Analyze infrastructure security posture and identify vulnerabilities

Parameters:

  • integration: Target cloud provider integration (required)
  • scope: Analysis scope (resource, vpc, region, account)
  • checks: Security checks to perform (cis, pci, hipaa, custom)
  • severity_threshold: Minimum severity to report
  • include_remediation: Include remediation recommendations

monitor_health

Monitor resource health, performance, and security metrics

Parameters:

  • integration: Target cloud provider integration (required)
  • resource_ids: List of resources to monitor
  • metrics: Specific metrics to retrieve
  • time_range: Time period for metrics
  • anomaly_detection: Enable anomaly detection
  • alert_threshold: Threshold for generating alerts

Installation

Prerequisites

  • A Unizo API key
  • An active Public Cloud (Infra) integration (Azure, AWS, Heroku, Google Cloud Platform)
  • Node.js v20 or higher

MCP Configuration

Here is an example configuration for setting up the Unizo Public Cloud (Infra) MCP server:

{
"mcpServers": {
"unizo-public-cloud": {
"command": "npx",
"args": [
"mcp-remote",
"http://api.unizo.ai/mcp/public-cloud",
"--allow-http",
"--header",
"apikey:${UNIZO_API_KEY}"
],
"env": {
"UNIZO_API_KEY": "your_api_key"
}
}
}
}

Client Setup

For detailed setup instructions with specific AI clients:

Environment Variables

The following environment variables are required:

Error Handling

All tools return errors in a consistent format:

{
"error": {
"code": "RESOURCE_NOT_FOUND",
"message": "Resource 'vm-abc123' not found"
}
}

Common Error Codes

  • INTEGRATION_NOT_FOUND: Invalid cloud integration ID
  • RESOURCE_NOT_FOUND: Specified resource not found or inaccessible
  • REGION_NOT_SUPPORTED: Region is not available for the integration
  • INVALID_CONFIGURATION: Provided configuration is invalid or unsupported
  • UNAUTHORIZED: Invalid API key or insufficient permissions
  • RATE_LIMIT_EXCEEDED: API quota exceeded