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.
/plugin marketplace add vinnie357/claude-skillsnpx agentmods add plugins/vinnie357/claude-skills/agent-sandboxinggit clone --depth 1 https://github.com/vinnie357/claude-skillsWrote 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/plugins/vinnie357/claude-skills/agent-sandboxing)<a href="https://agentmods.dev/plugins/vinnie357/claude-skills/agent-sandboxing"><img src="https://agentmods.dev/badge/plugins/vinnie357/claude-skills/agent-sandboxing.svg" alt="Measured on agentmods" height="20"></a>Grade A, and why
agent-sandboxing 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 yesterday.
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.
What it actually says
{
"name": "agent-sandboxing",
"version": "0.2.14",
"description": "Agent sandboxing for AI coding agents: microVM and kernel-isolated execution environments. Index skill (sandbox) routes to substrate-specific paths. Currently covers the kubernetes-sigs/agent-sandbox path on kind, GKE, and kina (Kubernetes-in-Apple-Containers), with Kata, gVisor, and Apple Container microVM substrates. Ships 7 skills + 6 slash commands + nushell scripts. Designed to expand to non-k8s sandboxing paths.",
"author": {
"name": "vinnie357"
},
"repository": "https://github.com/vinnie357/claude-skills",
"license": "MIT",
"keywords": [
"agent-sandbox",
"sandboxing",
"microvm",
"kata",
"gvisor",
"kina",
"apple-container",
"kubernetes",
"gke",
"kind",
"claude-code"
],
"skills": [
"./skills/sandbox",
"./skills/k8s-agent-sandbox",
"./skills/kata-on-kind",
"./skills/kata-on-gke",
"./skills/kina-microvm",
"./skills/claude-code-on-sandbox",
"./skills/agent-substrate-overview"
],
"commands": [
"./commands/bootstrap.md",
"./commands/provision.md",
"./commands/exec.md",
"./commands/reap.md",
"./commands/status.md",
"./commands/build-image.md"
]
}
What it installs
The manifest is a name and a version. 7 skills, 6 commands travel with it, and installing the plugin installs all of them — 781 tokens a session between them. Each is measured on its own page, and each can be installed alone.
- Skill agent-substrate-overview A 87 tokens
- Skill k8s-agent-sandbox A 89 tokens
- Skill kina-microvm A 89 tokens
- Skill sandbox A 88 tokens
- Skill kata-on-gke A 88 tokens
- Skill claude-code-on-sandbox B 99 tokens
- Skill kata-on-kind B 78 tokens
- Command bootstrap A 44 tokens
- Command provision A 23 tokens
- Command exec A 21 tokens
- Command status A 27 tokens
- Command reap A 23 tokens
- Command build-image C 25 tokens
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.
- yesterday First seen · 41 lines scan A bd0569452e8d
agent-sandboxing is a plugin published in the GitHub repository vinnie357/claude-skills (24 stars, last pushed today), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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 2026-09-03.
Other plugins, from other repositories
monk
Deploy and operate full applications with Monk DevOps Agent — cloud infra, SaaS integrations, and containerized workloads — from one chat. Bundles the monk skill and blocks accidental monk CLI shell-outs that would desync cluster state.
daytona
Creates and manages isolated cloud sandboxes (secure code execution environments with dedicated runtimes) on the Daytona platform.
open-forge
Self-host open-source apps on your own cloud, your own VPS, your own Kubernetes cluster, a PaaS, or your own laptop. Chat-friendly interface to existing tools (aws, az, hcloud, doctl, gcloud, flyctl, kubectl, helm, podman, docker, lume, tailscale, terraform, cdk, ssh, gh) — Claude runs every command; you only answer…
ensemble-infrastructure
Infrastructure management for AWS, Kubernetes, Docker, Helm, Fly.io.
agentic-workflows
Intelligent multi-agent swarms for automation - Ansible collections, Kubernetes operators, Terraform modules.
azure-containers
Azure Container Apps, Container Instances, and Container Registry — build, push, deploy, and scale containerized workloads.