AI & Automationaditi-p31/cache-divergence-study

llm-cachecheck

Measure how often the prompt cache changes greedy answers on your LLM server (llama.cpp, plus an experimental two-server mode for vLLM and SGLang)

Suggested install command

npx skills add aditi-p31/cache-divergence-study/llm-cachecheck

Always inspect the linked repository and skill instructions before running commands. Skills are instructions; permissions and execution still matter.

Compatibility

3 supported