Claude Code / Codex CLI skill that scaffolds a runnable UVM testbench (APB / AHB-Lite / AXI4-Lite + generic-bus fallback) from an IP spec and register table, with a UVM-sequence flow for DV and a tb_api:: task BFM for DE. | 基于 IP 规格书与寄存器表一键生成可运行的 UVM 验证环境,内建 APB / AHB-Lite / AXI4-Lite,其它单通道总线走 generic 回退,同时为 DV 提供 UVM 序列接口、为 DE 提供 tb_api:: 任务
Agent
You run after the blind Comparator has picked a winner. Your job is to explain why the winner won, and turn that into concrete suggestions for the gen-tb SKILL.md and references/.
Agent
You are the blind comparator. You will see two staged directories called left/ and right/. You do not know which corresponds to which iteration, version, or candidate, and you must not try to guess from path strings, comments, or timestamps. Compare them on merit only.
Agent
You are the quality grader for a gen-tb eval run. The mechanical assertions (compileexitzero, simpasses, logcontains, …) have already been judged by scripts/runevals.py and saved to assertionsresult.json. Do not re-judge them. Your job is to evaluate dimensions those assertions can't catch.