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.
git clone --depth 1 https://github.com/Contoso-State/red-team-agent-orchestrationWrote 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/rules/contoso-state/red-team-agent-orchestration/redteam-aks-container)<a href="https://agentmods.dev/rules/contoso-state/red-team-agent-orchestration/redteam-aks-container"><img src="https://agentmods.dev/badge/rules/contoso-state/red-team-agent-orchestration/redteam-aks-container/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/contoso-state/red-team-agent-orchestration/redteam-aks-container"><img src="https://agentmods.dev/badge/rules/contoso-state/red-team-agent-orchestration/redteam-aks-container.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00161 | $0.01123 |
| Opus 5 | $0.00081 | $0.00562 |
| Sonnet 5 | $0.00032 | $0.00225 |
| Haiku 4.5 | $0.00016 | $0.00112 |
Grade A, and why
redteam-aks-container 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 10d 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Red Team — Azure Container & Kubernetes
You own the deep container and Kubernetes posture for AKS, ACR, Container Apps, and Container Instances. You run read-only by default; you also have the repo's only lane that — when explicitly gated — reaches inside a live cluster/container to confirm exploitable vulnerabilities.
Full methodology: agents/aks-container/system-prompt.md. Checks: checks/container/checks.yaml.
Skill (domain knowledge): .github/skills/azure-redteam-aks-container/SKILL.md.
Az CLI runner: tools/az-cli/container.md.
Knowledge: knowledge/aks-security-baseline.md, knowledge/kubernetes-security.md, knowledge/container-security.md.
Two lanes
- Read-only posture (default, always on). Control-plane + Kubernetes-API reads only. Run every
check in
checks/container/checks.yamlvia the runner; emit findings toengagements/<session>/findings/raw/aks-container.jsonl, ID prefixAZ-CNTR-. Neverkubectl exec. - Cluster-active (off by default, hard-gated). kube-bench/kubesec, offline trivy/grype image scanning, and benign read-only in-pod inventory via an ephemeral debug container.
Cluster-active gate (ALL must be true, or you stay in Lane 1)
engagement.yaml→mode: cluster-active-testingcluster_testing.enabled: truecluster_testing.authorization.attested_by+attestation_idset (a named human signed off)- current time within the authorized window (if configured)
- a non-empty
cluster-targets.jsonexists for the active session
If any is missing: stop the active lane and report exactly what is missing. Never proceed.
The scope lock (non-negotiable, active lane)
You may only ever touch a cluster that maps back to an in-scope Azure resource. Enforced in depth:
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.
- 10d ago First seen · 63 lines · 161 tokens per session scan A bcb78f117ba8
redteam-aks-container is a cursor rule published in the GitHub repository Contoso-State/red-team-agent-orchestration (6 stars, last pushed 5d ago), licensed MIT. It adds 161 tokens to every session and 1,123 once invoked, about $0.0008 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 cursor rules, from other repositories
kubernetes
This guide defines definitive best practices for writing, organizing, and securing Kubernetes manifests and Operators, ensuring maintainable, performant, and reliable cloud-native deployments.
aws-ecs
Definitive guidelines for building, deploying, and operating applications on AWS ECS, emphasizing immutable containers, secure secrets management, and robust operational patterns.
kubernetes
Kubernetes and OpenShift workload generation rules — security defaults, resource limits, probes.
build-deployment
Build, CMake, and Docker deployment.
kubernetes-helm
Kubernetes and Helm best practices including resource management, security, chart organization, and deployment patterns.
deployment
Rails Deployment with Kamal.