- Home
- Skills
- DevOps & Cloud
- log analyzer kubernetes
log analyzer kubernetes
Diagnostic agent for Kubernetes and OpenShift logs, identifying root causes for pod failures and security leaks.
$1
Works with the AI Agents you already use
log analyzer kubernetes
Example session with this skill installed
Analyze this: Pod my-app-5f7d is in CrashLoopBackOff. Exit code 137. kubectl describe shows: Limits: cpu: 500m, memory: 256Mi. Events: FailedScheduling: 0/3 nodes are available: 3 Insufficient memory.
- Read your context and instructions
- Compiled the log analyzer kubernetes
I have analyzed the pod status and event logs for the my-app-5f7d resource. The primary failure is a resource exhaustion issue at the node level combined with container-level memory limits.
Analysis Report: Kubernetes Resource Failure
| Attribute | Details |
|---|---|
| Technology | Kubernetes (Generic/Unspecified) |
| Status | CrashLoopBackOff / FailedScheduling |
| Exit Code | 137 (OOMKilled) |
| Confidencialidad | MEDIO (Namespaces/Pod names visible) |
| Diagnosis | The pod is being terminated by the OOM Killer (Exit Code 137) because it exceeded its 256Mi memory limit. Simultaneously, new replicas cannot start because the cluster nodes have insufficient memory to satisfy the requests. |
Critical Findings
- OOMKilled (137): The application is attempting to allocate more than 256Mi of RAM.
Resource Starvation: The FailedScheduling event confirms no nodes have enough unallocated memory to host the pod.
Limits vs Requests: The current limits are too low for the application's runtime requirements, but the infrastructure is also at capacity.
Open Questions
- Does the pod have
resources.requestsdefined, or onlylimits? - Can you provide the log from the previous instance using
kubectl logs my-app-5f7d --previous? - Are there
ResourceQuotaobjects active in this namespace?
Next steps
- Review application memory usage and increase
resources.limits.memoryto at least 512Mi if the node capacity allows. - Check node pressure using
kubectl top nodesto verify if scaling the cluster is necessary. - Verify if a memory leak was introduced in the latest deployment.
Connects securely to your tools. The creator never sees your data.
About this skill
Cloud-native applications often fail with cryptic error codes or silent crashes that leave developers digging through thousands of lines of logs. This skill acts as a Level 2/3 platform support agent to parse logs and resource descriptions, pinpointing the root cause of failures across Kubernetes and OpenShift environments.
What it does
- Root cause identification parses exit codes like 137 (OOMKilled) or 127 (command not found) to distinguish between application, platform, and configuration errors.
- Security auditing identifies and flags sensitive data such as base64 secrets, ServiceAccount tokens, and cloud-provider ARNs that should not be in logs.
- Resource analysis evaluates
describe podoutputs to catchFailedMount,ResourceQuotaviolations, and taints/tolerations issues. - OpenShift specialization handles OCP-specific resources like Routes, DeploymentConfigs, and Security Context Constraints (SCC) validation errors.
- Cloud-provider diagnostics troubleshoots provider-specific auth issues including EKS IRSA, GKE KSA-GSA bindings, and AKS Workload Identity.
How it works
- Log ingestion where the user provides logs or
kubectl/occommand outputs. - Context gathering where the agent asks clarifying questions about the distribution (EKS, GKE, AKS, OCP), resource limits, or CNI plugins.
- Structured diagnosis which generates a standardized executive report identifying criticality, findings, and sensitive data risks.
Frameworks & tools
Kubernetes (vanilla, EKS, GKE, AKS) and OpenShift (OCP). Supports diagnostic data from kubectl and oc CLI tools, focusing on Pods, Deployments, Ingress, Routes, and SCCs.
Why this beats prompting it yourself
Generic prompts often miss the nuances of Kubernetes exit codes or the hierarchical nature of OpenShift SCCs. This skill enforces a structured diagnostic methodology that proactively checks for missing context like --previous logs and identifies sensitive data leaks that standard LLM prompts ignore.
Use cases
- Debugging
CrashLoopBackOffcycles by analyzing previous container logs and dependencies. - Resolving
Pendingpods by checkingResourceQuotaand scheduler constraints. - Fixing OpenShift permission issues related to
Security Context Constraints. - Auditing logs for exposed credentials, IPs, and internal cluster tokens.
Known limitations
This skill does not execute commands or modify live resources. It requires users to manually provide log and description outputs for analysis.
Requires manual log/describe output. Does not execute commands or modify cluster state. Analysis accuracy depends on providing resource limits and environment context.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 28 days ago
Permissions
No special permissions declared or detected