Recommendations API

The recommendations endpoint returns cost optimization suggestions generated by Kubeadapt's analysis engine. Each recommendation includes estimated savings and the current cost of the affected resource.

Required scope: recommendations:read

List Recommendations

text

GET /v1/recommendations

Returns a paginated list of recommendations across your organization.

Query Parameters

Parameter	Type	Default	Description
`cluster_id`	string		Filter by cluster ID
`recommendation_type`	string		Filter by recommendation type (e.g., `cpu_rightsizing`, `memory_rightsizing`, `idle_workload`)
`status`	string		Filter by status (e.g., `pending`, `applied`, `dismissed`)
`limit`	integer	`100`	Number of results to return (1 to 500)
`offset`	integer	`0`	Number of results to skip

Example Request

bash

curl -H "Authorization: Bearer ka_your_api_key" \
  "https://public-api.kubeadapt.io/v1/recommendations?cluster_id=cls-abc123&status=pending"

Example Response

json

{
  "recommendations": [
    {
      "id": "rec-001abc",
      "cluster_id": "cls-abc123",
      "cluster_name": "prod-us-east",
      "recommendation_type": "cpu_rightsizing",
      "namespace": "production",
      "resource_name": "api-server",
      "resource_type": "Deployment",
      "title": "Reduce CPU request for api-server",
      "description": "The api-server deployment has requested 1.0 CPU cores but consistently uses less than 0.35 cores. Reducing the request to 0.4 cores would free capacity and lower costs.",
      "estimated_hourly_savings": 0.42,
      "estimated_monthly_savings": 306.6,
      "current_hourly_cost": 0.85,
      "status": "pending",
      "created_at": "2026-03-10T08:00:00Z",
      "priority": "high"
    },
    {
      "id": "rec-002def",
      "cluster_id": "cls-abc123",
      "cluster_name": "prod-us-east",
      "recommendation_type": "idle_workload",
      "namespace": "non-production",
      "resource_name": "legacy-worker",
      "resource_type": "Deployment",
      "title": "Remove idle workload legacy-worker",
      "description": "The legacy-worker deployment has had zero CPU and memory usage for the past 14 days. Consider removing it if it is no longer needed.",
      "estimated_hourly_savings": 0.19,
      "estimated_monthly_savings": 138.7,
      "current_hourly_cost": 0.19,
      "status": "pending",
      "created_at": "2026-03-05T12:00:00Z",
      "priority": "medium"
    }
  ],
  "total": 5,
  "total_potential_savings_monthly": 1250.8
}

Response Fields

Each object in the recommendations array contains:

Field	Type	Description
`id`	string	Recommendation identifier
`cluster_id`	string	Cluster the recommendation applies to
`cluster_name`	string	Cluster display name
`recommendation_type`	string	Type of recommendation (e.g., `cpu_rightsizing`, `memory_rightsizing`, `idle_workload`)
`namespace`	string	Kubernetes namespace of the affected resource
`resource_name`	string	Name of the affected Kubernetes resource
`resource_type`	string	Kind of the affected resource (e.g., `Deployment`, `StatefulSet`)
`title`	string	Short description of the recommendation
`description`	string	Detailed explanation of the issue and suggested action
`estimated_hourly_savings`	float	Estimated hourly savings in USD if applied
`estimated_monthly_savings`	float	Estimated monthly savings in USD if applied
`current_hourly_cost`	float	Current hourly cost of the affected resource
`status`	string	Recommendation status: `pending`, `applied`, or `dismissed`
`created_at`	string	ISO 8601 timestamp when the recommendation was created
`priority`	string	Priority level: `high`, `medium`, or `low`. Computed from estimated monthly savings: >= $50 = `high`, >= $10 = `medium`, else `low`.

Top-Level Fields

Field	Type	Description
`total`	integer	Total number of matching recommendations
`total_potential_savings_monthly`	float	Sum of `estimated_monthly_savings` across all matching recommendations

Workloads - Metrics for recommended workloads
Clusters - Cluster-level cost data
Cost Allocation - Cost breakdown by team and department
Right-sizing - How right-sizing recommendations work