evalscope-cli
Installation
SKILL.md
EvalScope
Read only the relevant reference file for the matched workflow — don't preload all of them.
| Workflow | When | Reference |
|---|---|---|
| Eval (accuracy) | evaluate / benchmark / score | eval-reference.md |
| Perf (stress test) | throughput / latency / QPS / perf | perf-reference.md |
| RAG Evaluation | RAG / embedding / retrieval quality | rag-reference.md |
| Visualization | view results / compare / dashboard | (below) |
| Benchmark Discovery | list / find / what benchmarks | (below) |
| Troubleshooting | errors / failures / debug | troubleshooting.md |