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 Agent-Hellboy/mcp-runtime --skill k8s-hardening-auditgit clone --depth 1 https://github.com/Agent-Hellboy/mcp-runtimeWrote 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/agent-hellboy/mcp-runtime/k8s-hardening-audit)<a href="https://agentmods.dev/skills/agent-hellboy/mcp-runtime/k8s-hardening-audit"><img src="https://agentmods.dev/badge/skills/agent-hellboy/mcp-runtime/k8s-hardening-audit/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/skills/agent-hellboy/mcp-runtime/k8s-hardening-audit"><img src="https://agentmods.dev/badge/skills/agent-hellboy/mcp-runtime/k8s-hardening-audit.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 | $0.00109 | $0.02351 |
| Opus 5 | $0.00055 | $0.01175 |
| Sonnet 5 | $0.00022 | $0.00470 |
| Haiku 4.5 | $0.00011 | $0.00235 |
Grade A, and why
k8s-hardening-audit scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
wget -T2 -qO- http://mcp-platform-api.mcp-sentinel.svc:8080/health || echo BLOCKED How it starts
The opening of the file, as written. The whole thing — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Kubernetes Hardening Audit
Overview
Use this skill to assess MCP Runtime's cluster posture: RBAC graph, Pod
Security Standards, NetworkPolicy enforcement, manifest hygiene, and CIS
benchmark compliance. Findings use the shared template at
../_shared/FINDINGS-TEMPLATE.md.
For runtime authn/authz, gateway policy, and protocol fuzzing, use
security-audit-platform. For images and supply chain, use
supply-chain-audit.
Step 1 — Inventory
Repo-owned namespaces (per CLAUDE.md):
mcp-runtime— operator.mcp-sentinel— api, ui, ingest, processor, gateway, observability.mcp-servers— user MCP server workloads, gateway sidecars.registry— Distribution v2 registry.traefik— ingress controller (orkube-system/traefikon k3s).
For each:
NS=mcp-sentinel
kubectl -n "$NS" get all,sa,role,rolebinding,networkpolicy -o name
kubectl get clusterrole,clusterrolebinding -o name | grep -iE 'mcp|sentinel'
Record SA → Role/ClusterRole bindings as a graph; this becomes the input to RBAC analysis.
Step 2 — RBAC graph analysis
Install the helpers:
go install github.com/corneliusweig/rakkess/cmd/rakkess@latest
go install github.com/reactiveops/rbac-lookup@latest
Per ServiceAccount:
for sa in $(kubectl -n mcp-sentinel get sa -o name); do
echo "=== $sa ==="
rakkess --sa "$(echo $sa | cut -d/ -f2)" --namespace mcp-sentinel
done
Findings to flag:
*verbs onsecrets→ High unless explicitly justified by a single controller and scoped to a single namespace.- Cluster-scoped roles on workload SAs → High; scope to a Role unless the controller genuinely watches cluster-wide.
pods/exec,pods/portforward,pods/attach→ High outside break-glass paths.serviceaccounts/tokencreate → High; token impersonation primitive.escalateorbindon roles → Critical.- Bindings to
system:authenticatedorsystem:unauthenticated→ Critical.
Confirm Traefik narrowing (CLAUDE.md): bundled manifests should watch only
registry, mcp-sentinel, mcp-servers. Any extra namespace with a
traefik-watch Role binding without a documented reason is a Medium finding.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 238 lines · 109 tokens per session scan A 29f92193530a
k8s-hardening-audit is a skill published in the GitHub repository Agent-Hellboy/mcp-runtime (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 109 tokens to every session and 2,351 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
cloud-k8s
Use for authorized cloud, container, and Kubernetes security assessment including metadata SSRF, IAM misconfig, container escape paths, and cluster RBAC review.
competition-agent-cloud
Internal downstream skill for ctf-sandbox-orchestrator. CTF-sandbox workflow for AI-agent, prompt-injection, MCP or toolchain, cloud, container, CI/CD, and supply-chain challenges. Use when the user asks to analyze prompt-to-tool flows, retrieval poisoning, mounted secrets, deployment drift, runtime-vs-manifest…
competition-container-runtime
Internal downstream skill for ctf-sandbox-orchestrator. CTF-sandbox workflow for live container runtime analysis, mounted secrets, sidecars, namespaces, init containers, entrypoint drift, and route-to-container resolution. Use when the user asks why a live container differs from manifests, where a mounted secret is…
competition-kernel-container-escape
Internal downstream skill for ctf-sandbox-orchestrator. CTF-sandbox workflow for kernel attack surface, namespace and cgroup boundaries, container isolation assumptions, syscall paths, and escape primitive verification. Use when the user asks to analyze container-to-host escape paths, kernel exploit prerequisites…
competition-supply-chain
Internal downstream skill for ctf-sandbox-orchestrator. CTF-sandbox workflow for CI/CD, registry, dependency drift, artifact provenance, image build, release pipeline, and runtime consumer challenges. Use when the user asks to trace dependency drift, registry pulls, malicious packages, build or release tampering, CI…
deploy-docker-compose
Run the Omnigent server as a Docker compose stack (server + Postgres) on any Docker host — your laptop, a VPS, EC2 by hand, or as the base layer of any container-platform deploy. Invoke when the user wants to build the image, bring up the compose stack, debug the stack on a host they already have, or extend the stack…