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 shen-shanshan/vllm-dev-skills --skill vllm-test-generatorgit clone --depth 1 https://github.com/shen-shanshan/vllm-dev-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/skills/shen-shanshan/vllm-dev-skills/vllm-test-generator)<a href="https://agentmods.dev/skills/shen-shanshan/vllm-dev-skills/vllm-test-generator"><img src="https://agentmods.dev/badge/skills/shen-shanshan/vllm-dev-skills/vllm-test-generator/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/shen-shanshan/vllm-dev-skills/vllm-test-generator"><img src="https://agentmods.dev/badge/skills/shen-shanshan/vllm-dev-skills/vllm-test-generator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00135 | $0.01927 |
| Opus 5 | $0.00068 | $0.00963 |
| Sonnet 5 | $0.00027 | $0.00385 |
| Haiku 4.5 | $0.00014 | $0.00193 |
Grade A, and why
vllm-test-generator 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 11d 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 — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
vLLM Test Generator
Generate well-structured tests for vllm-project/vllm, following the conventions of the existing test suite.
Step 1 — Fetch Context
Before writing, fetch the relevant source and existing tests:
# Browse existing tests for the target area
gh api repos/vllm-project/vllm/contents/tests/<subdir> --jq '[.[] | {name}]'
# Read a representative existing test file
gh api repos/vllm-project/vllm/contents/tests/<path>.py --jq '.content' | base64 -d
# Read the source under test if needed
gh api repos/vllm-project/vllm/contents/vllm/<path>.py --jq '.content' | base64 -d
Step 2 — Classify the Test
| What is being tested | Type | Directory |
|---|---|---|
| Single function / class / method | Unit | tests/ or matching subdir |
| Config parsing, data structures, utils | Unit | tests/ root |
| CUDA kernels | Unit | tests/kernels/ |
| Attention backends | Unit/Integration | tests/v1/attention/ or tests/kernels/ |
| Full model inference with LLM class | Integration | tests/entrypoints/llm/ |
| OpenAI-compatible API | Integration | tests/entrypoints/openai/ |
| Model correctness (HF vs vLLM) | Integration | tests/basic_correctness/ or tests/models/language/ |
| Quantization end-to-end | Integration | tests/quantization/ |
| LoRA end-to-end | Integration | tests/lora/ |
| Distributed / multi-GPU | Integration | tests/distributed/ |
| v1 engine internals | Unit/Integration | tests/v1/ matching subdir |
| v1 e2e scenarios | Integration | tests/v1/e2e/general/ |
If the user specifies a directory, use it.
Step 3 — Write the Test
License header (required on every file)
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
Unit test pattern
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from unittest.mock import MagicMock, patch
import pytest
from vllm.<module> import <ClassName>
def test_<function_behavior>():
# Arrange
obj = <ClassName>(...)
# Act
result = obj.<method>(...)
# Assert
assert result == expected
@patch("vllm.<module>.<dependency>")
def test_<function_with_mock>(mock_dep):
mock_dep.return_value = ...
result = <function>(...)
mock_dep.assert_called_once_with(...)
assert result == expected
class Test<ClassName>:
def test_<method>(self):
...
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.
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.
- 11d ago First seen · 202 lines · 135 tokens per session scan A b297f1918e16
vllm-test-generator is a skill published in the GitHub repository shen-shanshan/vllm-dev-skills (17 stars, last pushed 2d ago), licensed Apache-2.0. It adds 135 tokens to every session and 1,927 once invoked, about $0.0007 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 2026-08-30.
Other skills, from other repositories
langchain4j-testing-strategies
Provides unit test, integration test, and mock AI patterns for LangChain4j applications. Creates mock LLM responses, tests retrieval chains, validates RAG workflows, and implements Testcontainers-based integration tests for Java AI services. Use when unit testing AI services, integration testing LangChain4j…
06-test
Write and iterate tests until they pass, or validate a user journey end to end in the browser. Use when the user wants to add coverage, find what's untested, or walk a flow. Not for auditing test health or debugging a failure.
knowledge-engineering-quality-and-delivery-unit-integration-and-shared-test-harnesses
A testing guide for the CLI and web interfaces, covering unit, integration, end-to-end, and real-process test setup with shared fixtures and cleanup.
archestra-dev-testing
Use when deciding whether a change needs a test and at which level — unit, backend route-level integration, MSW-backed frontend integration, or e2e — or when reviewing tests for the "fluff test" anti-pattern. Start here before archestra-dev-backend-tests or archestra-dev-e2e.
test-writing
Write comprehensive tests for code including unit tests, integration tests, and end-to-end tests. Use this to ensure code quality, catch bugs, and validate functionality.
extension-test
Set up and run unit, integration, and E2E tests for Chrome extensions. Covers Jest mocks for chrome. APIs and Puppeteer E2E with real Chrome.