Skip to main content

​
Overview

The Pastures Engine exposes a REST API that the Rancher extension consumes. All endpoints accept and return JSON (Content-Type: application/json).

​
Authentication

Authentication is optional and configured per-installation in Pastures → Settings. When an API key is set, every request includes an Authorization header: 
Authorization: Bearer <api-key>
If no key is configured, requests are sent without the header.

​
Base URL

The Engine URL is stored in localStorage and configurable in Settings. All API paths in this reference are relative to that base URL.

​
Endpoint Groups

Diagnostics

POST /api/diagnose — AI-powered root-cause analysis and remediation plans.

Advisories

GET /api/advisories — Security and operational advisories relevant to the cluster.

Health & Status

GET /health and GET /api/llm/status — Engine health check and AI provider status.

Clusters

GET /api/clusters/* — Cluster inventory, status, and metadata.

Operations

/api/cis/*, /api/etcd/*, /api/monitoring/*, /api/audit/* — CIS benchmarks, etcd health, monitoring, and audit logs.

GPU

/api/gpu/* — GPU inventory, utilization, and scheduling.

Support

POST /api/support-bundle — Generate and download support bundles.

Actions

POST /api/clusters/{id}/actions/execute — Execute remediation actions on a cluster.

​
Demo Mode

When demo mode is enabled in the extension, engineFetch() short-circuits and returns synthetic data from lib/demoResponses.ts without making any network calls. This allows the full UI to function without a running Engine instance.

​
Error Handling

The Engine returns standard HTTP status codes:
StatusMeaning
200Success
400Bad request — invalid or missing parameters
401Unauthorized — invalid or missing API key
404Endpoint not found
500Internal server error
503Service unavailable — Engine or AI provider is down
Error responses include a JSON body: 
{
  "error": "Description of what went wrong"
}