Skip to main content

serving-llms-vllm

Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throu

91/100

Security score

The serving-llms-vllm skill was audited on Feb 28, 2026 and we found 5 security issues across 2 threat categories. Review the findings below before installing.

Categories Tested

Security Issues

medium line 110

Curl to non-GitHub URL

SourceSKILL.md
110curl http://localhost:9090/metrics | grep vllm
low line 40

External URL reference

SourceSKILL.md
40client = OpenAI(base_url='http://localhost:8000/v1', api_key='EMPTY')
low line 100

External URL reference

SourceSKILL.md
100# Run: locust -f test_load.py --host http://localhost:8000
low line 110

External URL reference

SourceSKILL.md
110curl http://localhost:9090/metrics | grep vllm
low line 361

External URL reference

SourceSKILL.md
361- Community: https://discuss.vllm.ai
Scanned on Feb 28, 2026
View Security Dashboard