vllm

vllm is a skill for Claude Code, Codex from Prism-Shadow/penguin-harness. It costs 29 tokens per session (1,082 once invoked), scanned B, original, Apache-2.0.

A service for running open-weight language models on local NVIDIA or AMD graphics cards through an API that follows the OpenAI format. It includes tool-calling support for chat and agent applications.

In plain words
What is it for?
Use it to serve a selected language model, expose it to chat or coding agents, and connect applications through an OpenAI-compatible endpoint.
Why use it?
It provides a way to host a language model yourself instead of relying on a remote model service. It is aimed at workloads that need high-throughput responses and suitable GPU hardware.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to serve a selected language model, expose it to chat or coding agents, and connect applications through an OpenAI-compatible endpoint.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/prism-shadow/penguin-harness/vllm
About the project

PenguinHarness is a local-first platform in which multiple AI agents create, evaluate, optimize, and deploy agent applications. It is for people building AI software who want agents to generate applications and improve their own behavior through skills.

Prism-Shadow/penguin-harness · 2,025 stars · on GitHub · penguin.ooo

Install

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.

Any agent
npx skills add Prism-Shadow/penguin-harness --skill vllm
Clone the repo
git clone --depth 1 https://github.com/Prism-Shadow/penguin-harness

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for vllm

README.md
[![agentmods](https://agentmods.dev/badge/skills/prism-shadow/penguin-harness/vllm.svg)](https://agentmods.dev/skills/prism-shadow/penguin-harness/vllm)
Your own site
<a href="https://agentmods.dev/skills/prism-shadow/penguin-harness/vllm"><img src="https://agentmods.dev/badge/skills/prism-shadow/penguin-harness/vllm.svg" alt="Measured on agentmods" height="20"></a>
Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,082 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Excessive Agency · line 12
    Skill grants unrestricted tool access without appropriate constraints. An agent with unfettered tool access can perform arbitrary actions including file modification, network requests, and code execution.
    Fix: Restrict tool access to only the tools required for the skill's stated purpose. Use an explicit allowlist rather than granting blanket access.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00029 $0.01082
Opus 5 $0.00015 $0.00541
Sonnet 5 $0.00006 $0.00216
Haiku 4.5 $0.00003 $0.00108

Measured 3d ago against content hash f75268c10d50, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade B, and why

vllm scanned grade B with 2 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 3d 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.

Unrestricted tool accessmediumExcessive agency

A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.

If the user's message only invokes this skill (e.g. "use vllm skill") without a concrete request, ask the user what they want. Do not run any command until the goal is clear.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

4. Verify with `curl http://localhost:8000/v1/models`.
plugins/model-development/skills/vllm/SKILL.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

vLLM Serving

vLLM serves open-weight LLMs on local GPUs with high-throughput inference behind an OpenAI-compatible API, ready for chat and agent workloads.

Before you start

If the user's message only invokes this skill (e.g. "use vllm skill") without a concrete request, ask the user what they want. Do not run any command until the goal is clear.

Ask the user which model to serve; if they have no preference, recommend the small default Qwen/Qwen3.5-0.8B. Also ask what context length the workload needs.

vLLM needs an NVIDIA or AMD GPU. Engine choice follows the user's preference: Ollama also runs on GPUs and is the simpler default — pick vLLM for high-throughput serving, and Ollama on macOS or CPU-only machines, which vLLM does not serve. Confirm the hardware first:

nvidia-smi          # NVIDIA: GPU model and free VRAM (AMD ROCm: rocm-smi)
python3 --version   # a recent Python is required

The model must fit the available VRAM — model size and context length drive the serve flags below.

Suggested workflow

  1. Ask the user which model to serve; with no preference, recommend Qwen/Qwen3.5-0.8B.
  2. Pick the engine the user prefers: vLLM for high-throughput GPU serving; Ollama is the simple default and the choice on macOS or CPU-only machines.
  3. Serve on a free port, with the tool-calling flags whenever agents will call it (see below).
  4. Verify with curl http://localhost:8000/v1/models.
  5. Register the endpoint: penguin config model add ... --client-type openai-chat --base-url http://localhost:8000/v1 — a served model is not visible to Penguin until added.
  6. Confirm the new entry with penguin config model list.

Install

Use a fresh virtual environment (or uv):

python3 -m venv .venv && source .venv/bin/activate
pip install vllm

Serve

vllm serve Qwen/Qwen3.5-0.8B --port 8000

This exposes an OpenAI-compatible API at http://localhost:8000/v1. Key flags:

Read the full file on GitHub · 91 lines

Files

What ships with it

1 file 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.

Changes

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.

  1. 3d ago Changed · -4 lines f75268c10d50
  2. 9d ago First seen · 95 lines · 29 tokens per session scan B 44c2cc0fbe44

Subscribe to this mod's changes

vllm is a skill published in the GitHub repository Prism-Shadow/penguin-harness (2,025 stars, last pushed today), licensed Apache-2.0. It adds 29 tokens to every session and 1,082 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 2 findings (unrestricted tool access, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

LLM

Implement large language model (LLM) chat completions using the z-ai-web-dev-sdk. Use this skill when the user needs to build conversational AI applications, chatbots, AI assistants, or any text generation features. Supports multi-turn conversations, system prompts, and context management.

jjyaoao/HelloAgents · 59 tokens

ASCII Art

Generate ASCII art using pyfiglet (571 fonts), cowsay, boxes, toilet, image-to-ascii, remote APIs (asciified, ascii.co.uk), and LLM fallback. No API keys required.

agentic-in/elephant-agent · 47 tokens

data-leakage-detection

Detect sensitive information disclosure via escalating dialogue probes. Covers system prompt extraction, credential/API key leakage, PII, and internal configuration exposure.

Tencent/AI-Infra-Guard · 34 tokens

groq-inference

Ultra-fast LLM inference on custom LPU hardware. OpenAI-compatible API at api.groq.com. Lowest latency in the industry (500-1000+ tok/s). Supports chat completions, vision, audio (Whisper STT + TTS), tool calling, JSON mode, and streaming. Free tier available. Inference only — no training.

synthetic-sciences/openscience · 77 tokens

agentfield

Design and ship a multi-agent system on AgentField. Use when the user asks to build, scaffold, design, or run an agent, reasoner network, multi-agent backend, or 'an agent that does X' — whenever the work would otherwise be a single LLM call or a flat LangChain/CrewAI/AutoGen chain. The skill produces composite…

Agent-Field/agentfield · 127 tokens

data-fetching

Use when implementing or debugging ANY network request, API call, or data fetching. Covers fetch API, axios, React Query, SWR, error handling, caching strategies, offline support.

Intelligent-Internet/ii-agent · 41 tokens