Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/arjunprabhulal/devops-skills/gitopsnpx skills add arjunprabhulal/devops-skills --skill gitopsgit clone --depth 1 https://github.com/arjunprabhulal/devops-skillsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00122 | $0.01378 |
| Opus 5 | $0.00061 | $0.00689 |
| Sonnet 5 | $0.00024 | $0.00276 |
| Haiku 4.5 | $0.00012 | $0.00138 |
Grade A, and why
gitops scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GitOps
GitOps is not "we keep our YAML in Git." It is a specific operational model: a controller inside the cluster continuously pulls desired state from a repo and reconciles reality toward it, so the repo is not documentation of what was deployed — it is the only correct description of what should be deployed, and anything else is drift waiting to be corrected. That shift from push to pull is what makes the rest of the practice fall into place: audit trail, rollback, and promotion all become Git operations instead of pipeline scripts.
If the cluster state and the repo disagree, the cluster is wrong, not the repo.
1. Separate the app repo from the config repo
The repo where application code lives and the repo the reconciler watches should not be the same
repo, and should almost never be the same commit. CI in the app repo builds an image, pushes it to
a registry, and then makes a small, automated commit to the config repo bumping an image tag — it
never touches the cluster directly. This split matters because it lets you reason about "what is
deployed" by reading one small, low-churn repo instead of grepping application history, and it lets
the config repo have its own review rules (e.g. required approval for prod) independent of code
review norms. See ci-pipelines for the build side and artifact-management for where the image
actually lives.
Done when: deploying a new version never requires a human to run a deploy command by hand.
2. Model environments as directories or branches promoted by merge
Represent each environment (dev, staging, prod) as its own path or overlay, and promote a change by
merging or copying it forward — not by re-running a pipeline with a different target flag. A pull
request from staging/ into prod/ is a promotion event with a diff, a reviewer, and a timestamp,
which is a far stronger artifact than a Jenkins job log claiming the same thing happened. Kustomize
overlays or Helm values-per-environment both work; what matters is that the mechanism of promotion
is a Git operation everyone can see. See environment-management for how environments are defined
and release-management for gating promotion on approvals or criteria.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 99 lines · 122 tokens per session scan A bbd5d259da69
gitops is a skill published in the GitHub repository arjunprabhulal/devops-skills (2 stars, last pushed 7d ago), licensed MIT. It adds 122 tokens to every session and 1,378 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
google-mobile-ads-android-migrate-to-next-gen
Migrates Android applications from the old, legacy Google Mobile Ads (GMA) SDK (com.google.android.gms:play-services-ads) to the new GMA Next-Gen SDK (com.google.android.libraries.ads.mobile.sdk:ads-mobile-sdk). Provides comprehensive mapping tables for imports, classes, and method signatures to help determine…
agent-platform-inference
Connects to and performs inference with Google Cloud Agent Platform GenAI models, including First-Party Gemini models and Third-Party OpenMaaS models (Llama, DeepSeek, Qwen, etc.). Use when asked to perform inference, ask a model a question, run a test prompt, execute chat completions, or generate code for calling…
agent-platform-tuning
Agent Platform Model Tuning. Use when you need to fine-tune open models or Gemini models using Agent Platform infrastructure. Don't use for model training outside Agent Platform, model deployment to endpoints (use agent-platform-deploy), or managing serving endpoints (use agent-platform-endpoint-management).
agent-platform-deploy
Deploy open models or custom weights from Model Garden to Agent Platform endpoints, check the status of an in-progress deployment operation, or clean up resources by undeploying models and deleting endpoints. Use when asked to actively deploy a model, list the Model Garden CATALOG of available models, check if a…
agent-platform-eval-flywheel
Measures and improves the quality of AI models and agents on Google Cloud using the Eval Quality Flywheel methodology. Use when evaluating an agent or model, building an eval dataset, picking or writing evaluation metrics, analyzing failures, comparing results before and after a fix, or when guidance is needed on…
agent-platform-alert-configuration
Configures best-practice alerting policies for AI agents using OpenTelemetry (OTel) metrics, generating output as Terraform (.tf) configuration files. Use when analyzing, writing, or deploying alerting policies to monitor agent latency, error rates, token usage, and quality metrics. Don't use for standard…