تشغيل أي مهارة في Manus بنقرة واحدة

kubetbs

النجوم٢

التفرعات٠

آخر تحديث٢٨ مارس ٢٠٢٦ في ٠٦:٥٤

Troubleshoot a problematic microservice in Kubernetes by correlating events, logs and pods across namespaces. Usage: /kubetbs <key> <user> <host_ip> <namespace> <service>

التثبيت

التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.

تشغيل في Manus

المصدر

incognia

incognia/rules

فتح مستودع GitHub عرض مستودعات المنشئ

تنزيل

تشغيل في Manus

المهن ذات الصلةSOC

استنادا إلى تصنيف SOC المهني

مديرو الشبكات وأنظمة الحاسوبمهن الحاسوب والرياضيات·SOC 15-1244

SKILL.md

readonly

name	kubetbs
description	Troubleshoot a problematic microservice in Kubernetes by correlating events, logs and pods across namespaces. Usage: /kubetbs <key> <user> <host_ip> <namespace> <service>

Kubernetes troubleshooting

Arguments

Key: $0 (SSH key: incognia, faraday, kone, cad)
User: $1 (SSH user, e.g. ubuntu)
Host IP: $2 (server IP)
Namespace: $3 (namespace where the problem is)
Service: $4 (problematic service name, optional — if omitted, detect automatically)

Instructions

Read the full CoT: Load and follow ~/rules/cot/kubetbs.md from line 1 to end
Read the rules: Load ~/rules/rulesets/KUBETBS.md for methodology and common errors
Connect via SSH: ssh -i ~/.ssh/$0 $1@$2 — check ~/.ssh/config for custom port
Events: kubectl get events -n $3 --sort-by=.lastTimestamp | tail -30
Pods with restarts: kubectl get pods -n $3 --sort-by=.status.containerStatuses[0].restartCount | tail -20
If $4 was provided, focus on that service. Otherwise, pick the pod with most restarts.
Describe pod: kubectl describe pod -n $3 <pod> — identify owner and error
Logs by label: kubectl logs -n $3 -l app=$4 --since=60m --tail=500
Check Service/Endpoints: kubectl get svc,ep -n $3 | grep -E "$4|<dependency>"
Cross-namespace search: kubectl get svc -A | grep -i <dependency> if dependency not found locally
DNS verification: kubectl -n $3 run tmp --rm -it --image=busybox --restart=Never -- nslookup <dependency>.<namespace>.svc.cluster.local
Propose fix: describe correction WITHOUT deploying. Validate with temporary pod first.

Critical rule

Never deploy a fix without validating the hypothesis first. Use temporary busybox pod for DNS and connectivity checks.

Examples

/kubetbs kone ubuntu 30.30.40.90 sinaloa-qa ms-macroevento — troubleshoot specific service
/kubetbs kone ubuntu 30.30.40.90 michoacan-qa — auto-detect problematic service
/kubetbs cad ubuntu 10.0.1.50 produccion payments-api — client server

References

Detailed CoT: ~/rules/cot/kubetbs.md
Rules: ~/rules/rulesets/KUBETBS.md
Cluster analysis: /kube skill for general health first

المزيد من هذا المستودع

نفس المستودع

bmail

incognia/rules

Compose structured business-email drafts in English from curated templates and phrase banks from Business Email (book). Use this whenever the user asks for enquiry, informing, request, complaint, apology, order-status, or meeting-style business emails.

2026-06-222

aws-naming

incognia/rules

Normalize filenames for AWS S3 and CloudFront: lowercase, safe characters, extension normalization (.jpeg → .jpg), and .tsv audit log. Usage: /aws-naming <directory_or_file>

2026-06-012

styling

incognia/rules

Apply Kabat One Markdown styling to a file by platform. Usage: /styling <hedgedoc|gitlab|github> [mit|gpl] <file>

2026-05-252

changelogger

incognia/rules

Maintain CHANGELOG.md with correct CST dates, reverse chronological order, and Spanish Mexican language. Use this skill when updating the changelog.

2026-05-152

commit

incognia/rules

Full commit workflow with mandatory CHANGELOG validation, SSH validation, and conventional commits in English. Use this skill before making any git commit.

2026-05-152

git-init

incognia/rules

Initialize a git repo with SSH, identity and remote configured. Usage: /git-init <personal|laboral> <key_name> <remote_url> <branch>

2026-05-112

Kubernetes troubleshooting

Arguments

Key: $0 (SSH key: incognia, faraday, kone, cad)

User: $1 (SSH user, e.g. ubuntu)

Host IP: $2 (server IP)

Namespace: $3 (namespace where the problem is)

Service: $4 (problematic service name, optional — if omitted, detect automatically)

Instructions

Read the full CoT: Load and follow ~/rules/cot/kubetbs.md from line 1 to end

Read the rules: Load ~/rules/rulesets/KUBETBS.md for methodology and common errors

Connect via SSH: ssh -i ~/.ssh/$0 $1@$2 — check ~/.ssh/config for custom port

Events: kubectl get events -n $3 --sort-by=.lastTimestamp | tail -30

Pods with restarts: kubectl get pods -n $3 --sort-by=.status.containerStatuses[0].restartCount | tail -20

If $4 was provided, focus on that service. Otherwise, pick the pod with most restarts.

Describe pod: kubectl describe pod -n $3 <pod> — identify owner and error

Logs by label: kubectl logs -n $3 -l app=$4 --since=60m --tail=500

Check Service/Endpoints: kubectl get svc,ep -n $3 | grep -E "$4|<dependency>"

Cross-namespace search: kubectl get svc -A | grep -i <dependency> if dependency not found locally

DNS verification: kubectl -n $3 run tmp --rm -it --image=busybox --restart=Never -- nslookup <dependency>.<namespace>.svc.cluster.local

Propose fix: describe correction WITHOUT deploying. Validate with temporary pod first.

Critical rule

Never deploy a fix without validating the hypothesis first. Use temporary busybox pod for DNS and connectivity checks.

Examples

/kubetbs kone ubuntu 30.30.40.90 sinaloa-qa ms-macroevento — troubleshoot specific service

/kubetbs kone ubuntu 30.30.40.90 michoacan-qa — auto-detect problematic service

/kubetbs cad ubuntu 10.0.1.50 produccion payments-api — client server

References

Detailed CoT: ~/rules/cot/kubetbs.md

Rules: ~/rules/rulesets/KUBETBS.md

Cluster analysis: /kube skill for general health first