### GET /q/health/ready

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/trustyai-explainability.md

Readiness probe endpoint.

```APIDOC
## GET /q/health/ready

### Description
Readiness probe endpoint.

### Method
GET

### Endpoint
/q/health/ready

### Parameters
None

### Response
#### Success Response (200)
- **status** (string) - Health status (e.g., UP).

#### Response Example
```json
{
  "status": "UP"
}
```
```

--------------------------------

### GET /readyz

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/feast.md

Readiness probe for the Feast Operator manager.

```APIDOC
## GET /readyz

### Description
Returns the readiness status of the Feast Operator manager.

### Method
GET

### Endpoint
/readyz

### Response
#### Success Response (200)
- **status** (string) - "ok"

#### Response Example
{
  "status": "ok"
}
```

--------------------------------

### GET /info/inference/ids/{model}

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/trustyai-explainability.md

Get inference IDs for a specific model.

```APIDOC
## GET /info/inference/ids/{model}

### Description
Get inference IDs for a specific model.

### Method
GET

### Endpoint
/info/inference/ids/{model}

### Parameters
#### Path Parameters
- **model** (string) - Required - The name of the model.

#### Query Parameters
None

#### Request Body
None

### Response
#### Success Response (200)
- **inference_ids** (array) - A list of inference IDs for the specified model.

#### Response Example
```json
{
  "inference_ids": ["id1", "id2"]
}
```
```

--------------------------------

### API Server Main Application Logic (Go)

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/batch-gateway.md

The main.go file for the API server initializes the application, sets up configuration, and starts the HTTP server. It serves as the entry point for the API server component.

```go
package main

import (
	"context"
	"flag"
	"os"
	"os/signal"
	"syscall"

	"github.com/go-logr/logr"
	"github.com/opendatahub-io/architecture-context/internal/apiserver/config"
	"github.com/opendatahub-io/architecture-context/internal/apiserver/server"
	"github.com/opendatahub-io/architecture-context/internal/util/tls"
)

var ( 
	configPath = flag.String("config", "/etc/config/config.yaml", "Path to configuration file.")
)

func main() {
	log := logr.Discard()

	cfg, err := config.LoadConfig(*configPath)
	if err != nil {
		log.Error(err, "Failed to load configuration")
		os.Exit(1)
	}

	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

	// Setup signal catching
	sgo := make(chan os.Signal, 1)
	s// SIGINT (Ctrl+C) and SIGTERM (kill)
	ssignal.Notify(sgo, syscall.SIGINT, syscall.SIGTERM)

	// Initialize TLS if enabled
	tlsConfig, err := tls.LoadTLSConfig(cfg.TLS)
	if err != nil {
		log.Error(err, "Failed to load TLS configuration")
		os.Exit(1)
	}

	// Initialize the API server
	apiServer, err := server.NewServer(ctx, log, cfg, tslConfig)
	if err != nil {
		log.Error(err, "Failed to create API server")
		os.Exit(1)
	}

	// Start the API server in a goroutine
	go func() {
		log.Info("Starting API server on port ", cfg.Server.Port)
		if err := apiServer.ListenAndServe(); err != nil {
			log.Error(err, "API server failed")
			cancel()
		}
	}()

	// Wait for termination signal
	s<-sgo

	log.Info("Shutting down API server...")
	apiServer.Shutdown(ctx)
}

```

--------------------------------

### EPP Entry Point (cmd/epp/main.go)

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/llm-d-inference-scheduler.md

The main entry point for the EPP (presumably an inference engine or processing component). It initializes and starts the EPP service.

```go
package main

import (
	"context"
	"log"
	"os"

	"github.com/opendatahub-io/architecture-context/pkg/common"
	"github.com/opendatahub-io/architecture-context/pkg/epp"
)

func main() {
	ctx := context.Background()

	// Initialize EPP service
	eppService, err := epp.NewEPPService(ctx)
	if err != nil {
		log.Fatalf("failed to create EPP service: %v", err)
	}

	// Start EPP service
	if err := eppService.Run(ctx);
		err != nil {
		log.Fatalf("EPP service failed: %v", err)
		os.Exit(1)
	}
}

```

--------------------------------

### GET /api/v1/evaluations/jobs

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/diagrams/eval-hub-security-network.txt

Retrieves a list of evaluation jobs. This operation requires permissions to get evaluations.

```APIDOC
## GET /api/v1/evaluations/jobs

### Description
Retrieves a list of evaluation jobs. This operation requires permissions to get evaluations.

### Method
GET

### Endpoint
/api/v1/evaluations/jobs

### Parameters
#### Query Parameters
None

### Request Example
```json
{}
```

### Response
#### Success Response (200)
(Details of the success response are not provided in the source text, but would typically include a list of evaluation jobs)

#### Response Example
```json
[
  {
    "job_id": "job-1",
    "status": "completed"
  },
  {
    "job_id": "job-2",
    "status": "running"
  }
]
```

### RBAC Mapping
- API Group: trustyai.opendatahub.io, Resource: evaluations, Verb: get
```

--------------------------------

### GET /api/v1/evaluations/providers/{id}

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/eval-hub.md

Retrieves a specific evaluation provider. Use this endpoint to get details of a particular provider.

```APIDOC
## GET /api/v1/evaluations/providers/{id}

### Description
Retrieves a specific evaluation provider. Use this endpoint to get details of a particular provider.

### Method
GET

### Endpoint
/api/v1/evaluations/providers/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the evaluation provider.

#### Query Parameters
None

### Request Example
```json
{
  "example": "request body"
}
```

### Response
#### Success Response (200)
(Response details not specified in the provided text)

#### Response Example
```json
{
  "example": "response body"
}
```
```

--------------------------------

### Discover and analyze platform components via CLI

Source: https://github.com/opendatahub-io/architecture-context/blob/main/ARCHITECTURE-DIAGRAM-PROPOSAL.md

A command-line interface pattern for discovering platform components, cloning repositories, and triggering the aggregation process. It automates the workflow from repository discovery to final report generation.

```bash
/discover-platform-components --distribution=rhoai --version=3.3 --output-dir=~/rhoai-3.3-analysis
```

--------------------------------

### GET /api/v1/evaluations/collections/{id}

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/eval-hub.md

Retrieves a specific benchmark collection. Use this endpoint to get details of a particular collection.

```APIDOC
## GET /api/v1/evaluations/collections/{id}

### Description
Retrieves a specific benchmark collection. Use this endpoint to get details of a particular collection.

### Method
GET

### Endpoint
/api/v1/evaluations/collections/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the benchmark collection.

#### Query Parameters
None

### Request Example
```json
{
  "example": "request body"
}
```

### Response
#### Success Response (200)
(Response details not specified in the provided text)

#### Response Example
```json
{
  "example": "response body"
}
```
```

--------------------------------

### Controller Logic for Managing Gateway API Resources

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/prompts/openvino_model_server.md

This example demonstrates controller logic focused on managing Gateway API resources. It shows imports for Gateway API types and potentially the creation or reconciliation of `Gateway`, `HTTPRoute`, `GRPCRoute`, or `TLSRoute` resources. This is central to modern ingress management in Kubernetes.

```go
// Importing Gateway API types
import (
	gatewayv1 "gateway.networking.k8s.io/v1"
	// or gatewayv1beta1 "gateway.networking.k8s.io/v1beta1"
)

// Controller reconciliation loop for Gateway API resources
func (r *GatewayReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error) {
	// ... logic to manage Gateway, HTTPRoute, GRPCRoute, TLSRoute ...
	gateway := &gatewayv1.Gateway{}
	httpRoute := &gatewayv1.HTTPRoute{}
	// ... reconcile these resources ...

	return ctrl.Result{}, nil
}
```

--------------------------------

### GET /readyz

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/workload-variant-autoscaler.md

Readiness probe endpoint to verify the controller is initialized and ready to process requests.

```APIDOC
## GET /readyz

### Description
Readiness probe that checks if the controller has completed its bootstrap sequence, including ConfigMap loading.

### Method
GET

### Endpoint
/readyz

### Response
#### Success Response (200)
- **Body** (string) - "ok"
```

--------------------------------

### Kubernetes Network Policy Example

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/rhods-operator.md

Defines a Kubernetes NetworkPolicy to control network traffic to and from pods. This example specifically targets the kube-auth-proxy and restricts ingress and egress traffic, enhancing security.

```yaml
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
  name: kube-auth-proxy
  namespace: {{ .Release.Namespace }}
spec:
  podSelector:
    matchLabels:
      app: kube-auth-proxy
  policyTypes:
    - Ingress
    - Egress
  ingress:
    - from:
        - podSelector:
            matchLabels:
              app: ""
      ports:
        - protocol: TCP
          port: 8080
  egress:
    - to:
        - ipBlock:
            cidr: 10.0.0.0/8
    - to:
        - ipBlock:
            cidr: 172.16.0.0/12
    - to:
        - ipBlock:
            cidr: 192.168.0.0/16
```

--------------------------------

### Automate platform architecture documentation workflow

Source: https://github.com/opendatahub-io/architecture-context/blob/main/ARCHITECTURE-DIAGRAM-PROPOSAL.md

A shell-based workflow for cloning multiple component repositories, generating individual architecture summaries, and aggregating them into a single platform-level architecture document.

```bash
mkdir ~/rhoai-3.3-analysis
cd ~/rhoai-3.3-analysis

gh-org-clone --org red-hat-data-services \
  --output-dir ./repos \
  --branch release-3.3 \
  --repos kserve,odh-dashboard,data-science-pipelines,training-operator,model-registry

cd repos/kserve
/repo-to-architecture-summary --distribution=rhoai --version=3.3

cd ../odh-dashboard
/repo-to-architecture-summary --distribution=rhoai --version=3.3

cd ~/rhoai-3.3-analysis
/aggregate-platform-architecture --distribution=rhoai --version=3.3 --repos-dir=./repos

cat platform-architecture/rhoai-3.3-PLATFORM.md
```

--------------------------------

### Install Mermaid CLI

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/diagrams/distributed-workloads-README.md

Installs the Mermaid CLI globally using npm. This is a prerequisite for manually regenerating Mermaid diagrams into various image formats.

```bash
npm install -g @mermaid-js/mermaid-cli
```

--------------------------------

### Fetch Version-Specific Repositories and Generate Documentation

Source: https://github.com/opendatahub-io/architecture-context/blob/main/ARCHITECTURE-DIAGRAM-PROPOSAL.md

Automates the cloning of specific repository versions and running component agents to generate architecture summaries. This supports historical documentation and version-specific release tracking.

```bash
# Clone all opendatahub-io repos at v3.0 tag
gh-org-clone --org opendatahub-io \
  --output-dir ./repos/odh-v3.0 \
  --tag v3.0 \
  --include-archived=false

# Clone all red-hat-data-services repos at release-3.3 branch
gh-org-clone --org red-hat-data-services \
  --output-dir ./repos/rhoai-3.3 \
  --branch release-3.3 \
  --include-archived=false

# Run component agents on version-specific repos
./component-agent.sh --repo-dir ./repos/rhoai-3.3/kserve \
  --output ./summaries/rhoai-3.3/kserve.md

# Aggregate for specific version
./platform-agent.sh --summaries-dir ./summaries/rhoai-3.3 \
  --version 3.3 \
  --distribution rhoai \
  --output ./platform/rhoai-3.3-architecture.md
```

--------------------------------

### GET /q/health/live

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/trustyai-explainability.md

Liveness probe endpoint.

```APIDOC
## GET /q/health/live

### Description
Liveness probe endpoint.

### Method
GET

### Endpoint
/q/health/live

### Parameters
None

### Response
#### Success Response (200)
- **status** (string) - Health status (e.g., UP).

#### Response Example
```json
{
  "status": "UP"
}
```
```

--------------------------------

### Prepare opendatahub-operator Repository

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/prompts/kserve.md

Executes the 'get-manifests' make target for the opendatahub-operator repository. This clones kustomize manifests into the './opt/' directory, which are crucial for analyzing actual deployment configurations.

```bash
make get-manifests
```

--------------------------------

### Configure Environment and Execute Generation Pipeline

Source: https://context7.com/opendatahub-io/architecture-context/llms.txt

This snippet demonstrates how to initialize the environment variables from the template and run the architecture generation script. It supports both Anthropic and Vertex AI backends and requires the uv package manager to sync dependencies and execute the main Python application.

```bash
# Setup environment
cp .env.example .env

# Required for Anthropic API
ANTHROPIC_API_KEY=sk-ant-api03-...

# Alternative: Vertex AI credentials
GOOGLE_APPLICATION_CREDENTIALS=/path/to/service-account.json
VERTEX_AI_PROJECT=your-gcp-project
VERTEX_AI_LOCATION=us-central1

# Run with uv
uv sync
uv run python main.py all --platform=rhoai --branch=rhoai-3.4 --model=opus
```

--------------------------------

### GET /q/metrics

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/trustyai-explainability.md

Exposes Prometheus metrics (Micrometer).

```APIDOC
## GET /q/metrics

### Description
Exposes Prometheus metrics (Micrometer).

### Method
GET

### Endpoint
/q/metrics

### Parameters
None

### Response
#### Success Response (200)
- **metrics** (string) - Prometheus formatted metrics.

#### Response Example
```text
# HELP http_server_requests_seconds
# TYPE http_server_requests_seconds histogram
http_server_requests_seconds_bucket{exception="None",method="GET",uri="/info",status="200",le="0.005"} 0
...
```
```

--------------------------------

### Generate Product Architecture Summary (Shell)

Source: https://github.com/opendatahub-io/architecture-context/blob/main/ARCHITECTURE-DIAGRAM-PROPOSAL.md

Shell script to generate a simplified architecture summary for product documentation. It takes distribution, version, and output format as input, producing an output file.

```shell
#!/bin/bash
./platform-agent.sh --distribution rhoai --version 3.3 \
  --output-format product-docs \
  --output ./output/rhoai-3.3-product-arch.md

```

--------------------------------

### GET /metrics/all/requests

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/trustyai-explainability.md

List all scheduled metric requests.

```APIDOC
## GET /metrics/all/requests

### Description
List all scheduled metric requests.

### Method
GET

### Endpoint
/metrics/all/requests

### Parameters
None

### Response
#### Success Response (200)
- **scheduled_requests** (array) - A list of scheduled metric requests.

#### Response Example
```json
{
  "scheduled_requests": [
    {
      "id": "req1",
      "metric_type": "fairness",
      "status": "scheduled"
    }
  ]
}
```
```

--------------------------------

### Operator Entrypoint Script (entrypoint.sh)

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/spark-operator.md

This shell script serves as the entrypoint for the operator. It likely performs initial setup or configuration before executing the main operator process.

```shell
#!/bin/bash

set -e

# Start the controller manager
exec /usr/local/bin/manager "$@"

```

--------------------------------

### GET /q/metrics

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/trustyai-service-operator.md

Endpoint for Prometheus metrics scraping.

```APIDOC
## GET /q/metrics

### Description
Exposes service metrics in Prometheus format for monitoring and scraping.

### Method
GET

### Endpoint
/q/metrics

### Response
#### Success Response (200)
- **metrics** (text/plain) - Prometheus formatted metrics data.
```

--------------------------------

### GET /api/status

Source: https://github.com/opendatahub-io/architecture-context/blob/main/architecture/rhoai-3.4/odh-dashboard.md

Checks the user authentication status.

```APIDOC
## GET /api/status

### Description
Checks the user authentication status.

### Method
GET

### Endpoint
/api/status

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Request Example
None

### Response
#### Success Response (200)
- **authenticated** (boolean) - Indicates if the user is authenticated.

#### Response Example
{
  "authenticated": true
}
```