Lightspeed Prompts

Prompts by Category

Cluster Observability

1. `checkClusterHealth`

Check the overall health of the cluster with detailed information. Show node status, pod health, and any issues found.

2. `checkNodeConditions`

Check all node conditions in the cluster. Are there any nodes with memory pressure, disk pressure, PID pressure, or not ready?

3. `detectResourceIssues`

Detect resource issues in the cluster using CPU threshold 80%, memory threshold 85%, and restart threshold 5. Report any problematic pods.

4. `getPerformanceMetrics`

Get performance metrics for the openshift-lightspeed namespace. Show me CPU and memory usage for all pods.

5. `analyzePodDisruptions`

Analyze pod disruptions in namespace openshift-lightspeed for the last 24 hours. Were there any unexpected restarts, evictions, or OOM kills?

6. `checkKubeletStatus`

Check kubelet status on all cluster nodes. Look back 24 hours and include system errors. Show me any nodes where kubelet is unhealthy.

7. `checkCrioStatus`

Check CRI-O container runtime status on all nodes. Show recent logs and report any runtime failures.

8. `analyzeJournalctlPodErrors`

Analyze journalctl logs for pod-related errors across the cluster. Look for OOM kills, CrashLoopBackOff, and image pull failures.

9. `monitorDeployments`

Monitor all deployments in the openshift-lightspeed namespace. Show their rollout status, ready replicas, and any unavailable pods.

10. `monitorDeployments` (other namespace)

Check deployment status in the n8n-test namespace. Are all deployments healthy with all replicas ready?

Nodes & Metrics

11. `nodes_top`

Show me the current CPU and memory usage for all nodes in the cluster. Which node is under the most load?

12. `nodes_log`

Get the last 50 lines of kubelet logs from node ip-10-0-1-100. Are there any error messages?

13. `nodes_stats_summary`

Get detailed resource usage statistics for node ip-10-0-1-100 including CPU, memory, filesystem, network, and PSI metrics.

Pods

14. `pods_list_in_namespace`

List all pods in the openshift-lightspeed namespace. Show their status, node, and restart count.

15. `pods_list`

List all pods across the entire cluster. Filter by status to show only pods that are NOT in Running or Succeeded state.

16. `pods_get`

Get the full details of the pod openshift-mcp-server in namespace openshift-lightspeed. Show me its container images, resource limits, and environment variables.

17. `pods_log`

Get the last 100 lines of logs from the lightspeed-app-server pod in openshift-lightspeed namespace. Look for any errors or warnings.

18. `pods_top`

Show me the CPU and memory consumption of all pods in openshift-lightspeed namespace. Which pods are consuming the most resources?

19. `pods_exec`

Execute the command ["cat", "/etc/os-release"] in the openshift-mcp-server pod in namespace openshift-lightspeed. What OS is the container running?

20. `pods_run`

Run a temporary pod using the image busybox in namespace n8n-test. Name it debug-pod.

Namespaces & Projects

21. `namespaces_list`

List all namespaces in the cluster. Which ones were created by ArgoCD and which are system namespaces?

22. `projects_list`

List all OpenShift projects in the cluster. How many user projects vs system projects exist?

Generic Resources (resources_*)

23. `resources_list` — Deployments

List all deployments in the openshift-lightspeed namespace. Show their replicas and available counts.

24. `resources_list` — Services

List all services in namespace openshift-lightspeed. Show their types and cluster IPs.

25. `resources_list` — Routes

List all routes in namespace openshift-lightspeed. Show hostnames and TLS configuration.

26. `resources_list` — PVCs

List all PersistentVolumeClaims in namespace openshift-lightspeed. Show their status, capacity, and storage class.

27. `resources_list` — NetworkPolicies

List all NetworkPolicies in namespace openshift-lightspeed. Explain what traffic is allowed or blocked.

28. `resources_list` — ResourceQuotas

List all ResourceQuotas in namespace openshift-lightspeed. How close are we to reaching the CPU and memory limits?

29. `resources_list` — ConfigMaps

List all ConfigMaps in namespace openshift-lightspeed. How many are there and what are their names?

30. `resources_list` — Secrets

List all Secrets in namespace openshift-lightspeed. Show their types (Opaque, kubernetes.io/tls, etc).

31. `resources_list` — HPA

List all HorizontalPodAutoscalers in namespace openshift-lightspeed. Show current vs desired replicas and scaling metrics.

32. `resources_list` — Operator Subscriptions

List all Subscriptions in namespace openshift-operators with API group operators.coreos.com. Show installed CSV versions and update channels.

33. `resources_list` — MCPServerRegistrations

List all MCPServerRegistrations with API group mcp.kuadrant.io in namespace openshift-lightspeed. Show their Ready status and discovered tools count.

34. `resources_get` — Specific resource

Get the OLSConfig resource named "cluster" with apiVersion ols.openshift.io/v1alpha1. Show the configured MCP servers and LLM provider.

35. `resources_scale` — Scale

Scale the deployment n8n in namespace openshift-lightspeed to 2 replicas.

Events

36. `events_list`

List all events in the openshift-lightspeed namespace. Are there any Warning events related to image pull failures, OOM kills, or CrashLoopBackOff?

37. `events_list` (cluster-wide)

List all Warning events across all namespaces from the last few minutes. Summarize what went wrong.

Resource Provisioning

38. `createDeployment`

Create a new deployment called httpd-test in namespace n8n-test using image image-registry.openshift-image-registry.svc:5000/openshift/httpd:latest with 2 replicas.

39. `createService`

Create a ClusterIP service called httpd-test-svc in namespace n8n-test that targets pods with label app=httpd-test, exposing port 8080 to target port 8080.

40. `createHpa`

Create a HorizontalPodAutoscaler for deployment httpd-test in namespace n8n-test. Min 1 replica, max 5, scale at 70% CPU and 80% memory.

41. `createNetworkPolicy`

Create a deny-all NetworkPolicy in namespace n8n-test for pods with label app=httpd-test. Then create another allowing ingress only from pods with label app=n8n.

42. `deployDatabase`

Deploy a PostgreSQL database called test-db in namespace n8n-test for testing purposes.

43. `resources_create_or_update` — Direct YAML

Create a ConfigMap in namespace n8n-test with the following YAML:
apiVersion: v1
kind: ConfigMap
metadata:
  name: test-config
  namespace: n8n-test
data:
  APP_MODE: "production"
  LOG_LEVEL: "info"

Cleanup

44. `resources_delete`

Delete the deployment called httpd-test in namespace n8n-test. Then verify no pods remain.

45. `pods_delete`

Delete the pod openshift-mcp-server-0 in namespace openshift-lightspeed to force a restart. Verify the new pod comes up healthy.

Benchmarks

Note: These tools require kubectl in the MCP pod. Not available in Developer Sandbox.

46. `runCpuStressTest`

Run a CPU stress test of type load for 30 seconds using 2 cores and 256MB of memory. Show results.

47. `runStorageBenchmark`

Run a sequential-read storage benchmark with block size 4k for 30 seconds on a 1Gi volume using the default storage class.

48. `runNetworkTest`

Run a network bandwidth test using TCP protocol for 30 seconds with 2 parallel streams. Measure throughput.

49. `runDatabaseBenchmark`

Run a PostgreSQL oltp_read_write benchmark with 4 threads for 60 seconds on a table with 100000 rows.

50. `runKubeBurner`

Run a cluster-density-v2 kube-burner test with 5 iterations in namespace kube-burner-test. Create resources and then clean up.

Multi-tool Prompts

51. Complete namespace analysis

Give me a complete health report for the openshift-lightspeed namespace:
1. List all running pods and their resource usage
2. Check for any Warning events
3. Monitor deployment rollout status
4. List all routes and services

52. Problem diagnosis

Something seems wrong in the cluster. Check node conditions, detect resource issues with default thresholds, list recent Warning events across all namespaces, and analyze pod disruptions. Summarize findings.

53. Cluster inventory

Give me a full cluster inventory:
1. How many namespaces/projects exist?
2. List all nodes with their CPU/memory usage
3. Check overall cluster health
4. List all Operator Subscriptions and their status

54. Security audit

Perform a security audit of namespace openshift-lightspeed:
1. List all NetworkPolicies
2. List all Secrets and their types
3. Check if any pods are running as root
4. List all ServiceAccounts

55. Complete app provisioning

In namespace n8n-test:
1. Create a deployment called myapp using image nginx:latest with 2 replicas
2. Create a ClusterIP service for it on port 80
3. Create a HPA with min 1 max 5 replicas at 70% CPU
4. Verify everything is running

56. MCP Gateway status

Check the MCP Gateway infrastructure:
1. List MCPServerRegistrations in openshift-lightspeed namespace (group mcp.kuadrant.io)
2. List MCPGatewayExtensions in mcp-system namespace (group mcp.kuadrant.io)
3. List pods in mcp-system namespace
4. Are all components healthy?

Lightspeed Prompts

Prompts by Category

Cluster Observability

1. checkClusterHealth

2. checkNodeConditions

3. detectResourceIssues

4. getPerformanceMetrics

5. analyzePodDisruptions

6. checkKubeletStatus

7. checkCrioStatus

8. analyzeJournalctlPodErrors

9. monitorDeployments

10. monitorDeployments (other namespace)

Nodes & Metrics

11. nodes_top

12. nodes_log

13. nodes_stats_summary

Pods

14. pods_list_in_namespace

15. pods_list

16. pods_get

17. pods_log

18. pods_top

19. pods_exec

20. pods_run

Namespaces & Projects

21. namespaces_list

22. projects_list

Generic Resources (resources_*)

23. resources_list — Deployments

24. resources_list — Services

25. resources_list — Routes

26. resources_list — PVCs

27. resources_list — NetworkPolicies

28. resources_list — ResourceQuotas

29. resources_list — ConfigMaps

30. resources_list — Secrets

31. resources_list — HPA

32. resources_list — Operator Subscriptions

33. resources_list — MCPServerRegistrations

34. resources_get — Specific resource

35. resources_scale — Scale

Events

36. events_list

37. events_list (cluster-wide)

Resource Provisioning

38. createDeployment

39. createService

40. createHpa

41. createNetworkPolicy

42. deployDatabase

43. resources_create_or_update — Direct YAML

Cleanup

44. resources_delete

45. pods_delete

Benchmarks

46. runCpuStressTest

47. runStorageBenchmark

48. runNetworkTest

49. runDatabaseBenchmark

50. runKubeBurner

Multi-tool Prompts

51. Complete namespace analysis

52. Problem diagnosis

53. Cluster inventory

54. Security audit

55. Complete app provisioning

56. MCP Gateway status

1. `checkClusterHealth`

2. `checkNodeConditions`

3. `detectResourceIssues`

4. `getPerformanceMetrics`

5. `analyzePodDisruptions`

6. `checkKubeletStatus`

7. `checkCrioStatus`

8. `analyzeJournalctlPodErrors`

9. `monitorDeployments`

10. `monitorDeployments` (other namespace)

11. `nodes_top`

12. `nodes_log`

13. `nodes_stats_summary`

14. `pods_list_in_namespace`

15. `pods_list`

16. `pods_get`

17. `pods_log`

18. `pods_top`

19. `pods_exec`

20. `pods_run`

21. `namespaces_list`

22. `projects_list`

23. `resources_list` — Deployments

24. `resources_list` — Services

25. `resources_list` — Routes

26. `resources_list` — PVCs

27. `resources_list` — NetworkPolicies

28. `resources_list` — ResourceQuotas

29. `resources_list` — ConfigMaps

30. `resources_list` — Secrets

31. `resources_list` — HPA

32. `resources_list` — Operator Subscriptions

33. `resources_list` — MCPServerRegistrations

34. `resources_get` — Specific resource

35. `resources_scale` — Scale

36. `events_list`

37. `events_list` (cluster-wide)

38. `createDeployment`

39. `createService`

40. `createHpa`

41. `createNetworkPolicy`

42. `deployDatabase`

43. `resources_create_or_update` — Direct YAML

44. `resources_delete`

45. `pods_delete`

46. `runCpuStressTest`

47. `runStorageBenchmark`

48. `runNetworkTest`

49. `runDatabaseBenchmark`

50. `runKubeBurner`