How do I install the Ceet Devops Sre skill?

Run npx skillmds@latest add cmolg/ceet-devops-sre in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Ceet Devops Sre safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Ceet Devops Sre?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Ceet Devops Sre free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Ceet Devops Sre?

CMolG (@cmolg) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Ceet Devops Sre by cmolg — Agent Skill

Name: Ceet Devops Sre
Author: CMolG

Ceet Devops Sre

Cognitive Extraction Engine for DevOps and SRE practitioners. Use when a user wants to clone the judgment of an infrastructure, platform, or reliability engineer — how they think about CI/CD, observability, incident response, capacity planning, on-call, and reliability tradeoffs. Trigger on phrases like "extract my SRE thinking", "clone our platform engineer", "build an AI ops assistant that thinks like our oncall", "interview me about infrastructure", "capture how I run postmortems", or any request to distill DevOps/SRE judgment into a reusable AI configuration. Outputs are tool-agnostic.

CMolG 1c656d5 35 files · 32.2 KB Updated Sep 17, 2026

File contents

CEET — DevOps / SRE

What this skill does

Interview a platform / infra / reliability engineer. Produce two tool-agnostic artifacts: a cognitive clone of how they think about infrastructure and reliability, and an AI environment that applies their judgment in any AI tool.

When to use

The user wants an AI that reviews Terraform/Kubernetes/Helm/pipeline code with their specific standards.
A team wants a reusable "how our platform team thinks" artifact for AI assistants.
The user is codifying incident-response and postmortem voice.

Run order

Read interview/questions.md. Work theme by theme.
Extract. Tag: #mental-model, #heuristic, #procedure, #quality, #anti-pattern, #vocabulary.
Ask for incident stories. SRE judgment is overwhelmingly shaped by incidents — get 2-3 concrete ones.
Synthesize into the role-specific templates/cognitive-profile.md.
Activate into the role-specific templates/ directory. Save to templates/.
Hand off via ../docs/tool-integration.md.

SRE-specific things to listen for

Error-budget thinking. Do they run on SLOs? How do they negotiate with product?
Blast radius. When changing infra, how do they bound the worst case?
Toil detection. What do they consider toil, and what's their policy on automating vs. accepting it?
On-call philosophy. What deserves a page vs. a ticket? What shame-free practices do they enforce?
Postmortem voice. Blameless, narrative, action-item-driven — and what they refuse to include.
Observability layering. Logs / metrics / traces / profiles — when each, what they instrument first.
Change management. Canary / blue-green / feature flags / progressive delivery — their defaults.
IaC taste. Terraform / Pulumi / Crossplane / CDK — opinions and reasons.
Cost vs. reliability tradeoffs. When they push back on product, when they absorb cost.
Security reflexes. Least-privilege instincts, secrets handling, what they refuse to review.

Synthesis guardrails

Include at least one runbook-style decision procedure (e.g. "When latency SLO burns, I…").
Capture their page / no-page criteria explicitly — this shapes the AI's alerting advice.
Keep specific vendor/tool names if they matter (they often do in this role).
Save outputs to templates/.

---name: ceet-devops-sredescription: Cognitive Extraction Engine for DevOps and SRE practitioners. Use when a user wants to clone the judgment of an infrastructure, platform, or reliability engineer — how they think about CI/CD, observability, incident response, capacity planning, on-call, and reliability tradeoffs. Trigger on phrases like "extract my SRE thinking", "clone our platform engineer", "build an AI ops assistant that thinks like our oncall", "interview me about infrastructure", "capture how I run postmortems", or any request to distill DevOps/SRE judgment into a reusable AI configuration. Outputs are tool-agnostic.---# CEET — DevOps / SRE## What this skill doesInterview a platform / infra / reliability engineer. Produce two tool-agnostic artifacts: a **cognitive clone** of how they think about infrastructure and reliability, and an **AI environment** that applies their judgment in any AI tool.## When to use- The user wants an AI that reviews Terraform/Kubernetes/Helm/pipeline code with their specific standards.- A team wants a reusable "how our platform team thinks" artifact for AI assistants.- The user is codifying incident-response and postmortem voice.## Run order1. **Read** `interview/questions.md`. Work theme by theme.2. **Extract.** Tag: `#mental-model`, `#heuristic`, `#procedure`, `#quality`, `#anti-pattern`, `#vocabulary`.3. **Ask for incident stories.** SRE judgment is overwhelmingly shaped by incidents — get 2-3 concrete ones.4. **Synthesize** into the role-specific `templates/cognitive-profile.md`.5. **Activate** into the role-specific `templates/` directory. Save to `templates/`.6. **Hand off** via [`../docs/tool-integration.md`](../docs/tool-integration.md).## SRE-specific things to listen for- **Error-budget thinking.** Do they run on SLOs? How do they negotiate with product?- **Blast radius.** When changing infra, how do they bound the worst case?- **Toil detection.** What do they consider toil, and what's their policy on automating vs. accepting it?- **On-call philosophy.** What deserves a page vs. a ticket? What shame-free practices do they enforce?- **Postmortem voice.** Blameless, narrative, action-item-driven — and what they refuse to include.- **Observability layering.** Logs / metrics / traces / profiles — when each, what they instrument first.- **Change management.** Canary / blue-green / feature flags / progressive delivery — their defaults.- **IaC taste.** Terraform / Pulumi / Crossplane / CDK — opinions and reasons.- **Cost vs. reliability tradeoffs.** When they push back on product, when they absorb cost.- **Security reflexes.** Least-privilege instincts, secrets handling, what they refuse to review.## Synthesis guardrails- Include at least one **runbook-style decision procedure** (e.g. "When latency SLO burns, I…").- Capture their **page / no-page criteria** explicitly — this shapes the AI's alerting advice.- Keep specific vendor/tool names if they matter (they often do in this role).- Save outputs to `templates/`.

Frequently asked questions