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 skills add PrabhdeepSingh/claude-plugins --skill infra-standardsgit clone --depth 1 https://github.com/PrabhdeepSingh/claude-pluginsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/prabhdeepsingh/claude-plugins/infra-standards)<a href="https://agentmods.dev/skills/prabhdeepsingh/claude-plugins/infra-standards"><img src="https://agentmods.dev/badge/skills/prabhdeepsingh/claude-plugins/infra-standards.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00106 | $0.02804 |
| Opus 5 | $0.00053 | $0.01402 |
| Sonnet 5 | $0.00021 | $0.00561 |
| Haiku 4.5 | $0.00011 | $0.00280 |
Grade A, and why
infra-standards 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 7d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Infra standards — the outage-shaped mistakes are all preventable at review time
Infrastructure changes fail differently from code changes: there's no unit test that catches a deleted database, and the blast radius of one line can be the whole environment. These rules exist because every one of them maps to a class of incident — drift nobody can reproduce, a plan nobody read, a secret in an image layer, a pipeline greened by disabling its checks. The theme throughout: infrastructure is code — reviewed, planned, least-privileged, and boring.
How to apply this
When touching any infra surface, hold three questions: what is the blast radius of this change (what can it delete or replace), where do the secrets live (never in the artifact), and would this survive being run twice (idempotency). Run the self-check before shipping.
1. Everything is code — no clickops
Any change made by hand in a cloud console/portal is drift: invisible to review, unreproducible, and silently overwritten (or silently load-bearing) the next time the IaC applies. The rules:
- Infrastructure changes go through the IaC files in the repo, PR-reviewed like any code.
- A genuine emergency hand-fix is allowed — and codifying it is part of the same incident, not a follow-up ticket that dies in the backlog. Until it's in code, the fix doesn't exist.
- If IaC and reality disagree, reality is the bug: reconcile toward the code (import the resource or fix the code), never by editing the console to match.
2. Read the plan before you apply — every time
terraform plan / az deployment ... what-if / CloudFormation change sets exist because IaC's most dangerous property is that a one-line diff can mean "destroy and recreate." Changing an immutable attribute (a resource name, an AZ, a SKU) doesn't edit the resource — it replaces it, and for a database that's data loss wearing a green checkmark.
- Never apply a plan you didn't read. In the plan output,
destroyandreplacelines are the whole review — explain each one or stop. - In CI, the plan is posted for human review before apply on protected environments; auto-apply is for dev sandboxes only.
- State is sacred (Terraform): remote backend with locking, never hand-edited, never committed to the repo. Use
terraform statecommands (orimport/movedblocks) for surgery — a corrupted state file makes the plan lie to you.
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.
- 7d ago First seen · 135 lines · 106 tokens per session scan A a220b44fb6b5
infra-standards is a skill published in the GitHub repository PrabhdeepSingh/claude-plugins (3 stars, last pushed yesterday), licensed MIT. It adds 106 tokens to every session and 2,804 once invoked, about $0.0005 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
operating-infra
Author, inspect, troubleshoot, and review infrastructure across IaC, Kubernetes, cloud resources, containers, CI/CD, and Linux hosts. Use when changing Terraform/OpenTofu, Kubernetes, Helm, Kustomize, Dockerfiles, GitHub Actions workflow/job/permissions semantics, AWS, GCP, Cloud Run, BigQuery, IAM, logs, instances…
openshift-app
Package applications for OpenShift deployment: container images (UBI, arbitrary UID, multi-stage builds), packaging formats (Helm, Kustomize, Operators, OLM v1), CI/CD (Tekton, ArgoCD, Shipwright, Conforma), security (SCC, PSA, supply chain, image signing, secrets), operations (Routes, probes, scaling, monitoring…
ccc-devops
For scheduled deployments and automated infra tasks, see: https://code.claude.com/docs/en/scheduled-tasks — ccc-devops skills (especially cicd-pipeline-generator and zero-downtime-deploy) can be wrapped as scheduled routines for nightly deployments, weekly security scans, or periodic drift detection.
ci-cd-pipeline
Use near the end of a project to generate CI and make the app deploy-ready. Produces a polyglot GitHub Actions workflow (lint, typecheck, test, build) and multi-stage Dockerfiles. The finish line is CI-green and container-ready, not an actual deploy.
devops-patterns
DevOps patterns for infrastructure, CI/CD, and deployment automation. Use when configuring Docker containers, CI/CD pipelines, cloud deployments, Kubernetes, and monitoring. Covers GitHub Actions, Docker, Vercel, Railway, AWS, Terraform, and observability.
infrastructure-planning
Plan and document the required software, infrastructure, and logistics for a Paperclip company — domains, GitHub, Docker Hub, Kubernetes, Slack, Google Workspace, Stripe, shipping providers, and CI/CD pipelines. Use when setting up infrastructure for a new company or auditing what an existing company needs.