MCP / Skill Cross Benchmark
A reproducible HDL understanding benchmark comparing source text without skills, local pyslang-mcp without skills, and local pyslang-mcp with the rtl-lint-auditor skill context. Generated 2026-04-26T19:16:01+00:00.
Accuracy By Category
Cross Statistics
| Case | Category | Difficulty | Expected | Text/no skill | MCP/no skill | MCP/skill | Skill delta | Tokens | MCP/skill ms |
|---|