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 thevibeworks/claude-code-docs --skill setupgit clone --depth 1 https://github.com/thevibeworks/claude-code-docsWrote 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/thevibeworks/claude-code-docs/setup)<a href="https://agentmods.dev/skills/thevibeworks/claude-code-docs/setup"><img src="https://agentmods.dev/badge/skills/thevibeworks/claude-code-docs/setup/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/thevibeworks/claude-code-docs/setup"><img src="https://agentmods.dev/badge/skills/thevibeworks/claude-code-docs/setup.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 | $0.00087 | $0.04519 |
| Opus 5.5 | $0.00035 | $0.01808 |
| Sonnet 5.5 | $0.00017 | $0.00904 |
| Haiku 4.5 | $0.00009 | $0.00452 |
Grade A, and why
openshell-sandbox 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 14d 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.
curl -LsSf https://raw.githubusercontent.com/NVIDIA/OpenShell/v0.1.2/install.sh | OPENSHELL_VERSION=v0.1.2 sh This is a copy
100% identical to openshell-sandbox — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 372 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Setup walkthrough: NVIDIA OpenShell as a sandbox for Claude Managed Agents
Walk the user through these steps in order, from this directory. Each step
says what to run, what a healthy result looks like, and what to do when it is
not. Finish a step before you start the next. README.md has
the design and ../CLAUDE.md has the full debugging table.
Ground rules
- Never print the environment key or a session token. Ask the user to
paste the key into
.envthemselves. - Ask before you install anything. The OpenShell installer adds a system package and starts a service.
- Do not loosen the policy to make an error go away. Three of the five
commands in step 6 are supposed to fail. Keep
enforcement: enforceandcompatibility: hard_requirement: the defaults for both let violations through. - Shell variables do not outlive a shell. Steps 6 to 8 share
$session,$environment, and$sandbox. A person keeps one terminal open. If you are Claude Code, each command runs in a new shell, so start every command that uses them with the "set the variables again" lines in step 6, and run./start.shin the background with its output in a file you can read. - Look fields up, do not guess them. The policy reference at the end lists where each one is documented.
What you are building
your machine Anthropic
┌────────────────────────────────────────┐ ┌──────────────────────┐
│ start.sh │ │ self-hosted │
│ └─ ant beta:worker poll ──────────────┼─────▶│ environment │
│ │ environment key │ │ (a work queue) │
│ └─ on-work.sh, once per work item│ │ │
│ │ session token, on stdin │ │ agent loop │
│ ┌─────────▼──────────────────────┐ │ │ │
│ │ OpenShell sandbox, one/session │ │ │ │
│ │ ant beta:worker run ──────────┼────┼─────▶│ six routes: 4 on its │
│ │ └─ the agent's bash, files │ │ │ session, 2 on work │
│ └────────────────────────────────┘ │ │ items in its env │
│ policy.yaml, enforced outside it │ └──────────────────────┘
└────────────────────────────────────────┘
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.
- 14d ago First seen · 372 lines · 87 tokens per session scan A 627aef7cd8c5
openshell-sandbox is a skill published in the GitHub repository thevibeworks/claude-code-docs (150 stars, last pushed today), licensed MIT. It adds 87 tokens to every session and 4,519 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). It is 100% identical to openshell-sandbox, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
docker-management
Manage Docker containers, images, volumes, and Compose.
memstack-deployment-docker-setup
Use this skill when the user says 'Docker', 'Dockerfile', 'docker-compose', 'containerize', 'docker-setup', or needs to containerize an application with optimized Docker images and compose configurations. Do NOT use for serverless or static site deployments.
atlas-report
Render agent findings as a styled HTML report in the browser. Use when asked for "full report", "detailed report", "show in browser", or when CLI output exceeds the 40-line budget.
atlas-map
Map the system architecture — read the codebase, identify services and connections, output a C4-level architecture map as Mermaid diagrams with component descriptions. Use when asked to "map the architecture", "system diagram", "how does this work", or "architecture overview".
atlas-changelog
Maintain per-repo and cross-repo changelogs — append structured entries after agent work. Use when asked to "log this change", "update changelog", "what changed", "change history".
kubernetes-ops
Kubernetes and Helm in practice: Deployments, Services, probes, resources, autoscaling, disruption budgets, NetworkPolicy, Pod Security Standards, Gateway API (ingress-nginx is retired), Helm 4 charts and releases, and kubectl without accidents. ECS and EKS specifics go to AWS's aws-containers skill, Flux to…