Fini be04c8beea fix(ab-corpus): pin codex reasoning effort to medium ... gpt-5.5 在 codex CLI 默认走 xhigh，单调用 1-2 分钟，把 ab-v5 全 量 sweep 从 ~15 分钟拉到预测 13 小时。pin 到 medium 跟 ab-v4 gpt-5.4 历史值同档，保留 AB_CORPUS_CODEX_REASONING env override 留给以后想测全火力上限。		2026-04-29 09:49:59 +08:00
..
__tests__	fix(ab-corpus): preserve markdown structure across kept @domain blocks	2026-04-29 09:49:50 +08:00
clients	fix(ab-corpus): pin codex reasoning effort to medium	2026-04-29 09:49:59 +08:00
apply.ts	fix(ab-corpus): keep applying composite tags past per-shape failures	2026-04-29 09:49:55 +08:00
build-prompt.ts	fix(ab-corpus): preserve markdown structure across kept @domain blocks	2026-04-29 09:49:50 +08:00
measure-prompt-sizes.ts	feat(ab-corpus): per-domain cookbook filter (Phase 2 of token diet)	2026-04-29 09:49:49 +08:00
real-model.ts	feat(ab-corpus): per-domain cookbook filter (Phase 2 of token diet)	2026-04-29 09:49:49 +08:00
run.ts	feat(ab-corpus): bootstrap ab-v3 with token cost + composite difficulty	2026-04-29 07:45:00 +08:00
stub-model.ts	fix(ai-skills): drop invalid section_header subtitle from cookbook + stubs	2026-04-29 09:49:40 +08:00
write-report.ts	feat(ab-corpus): bootstrap ab-v3 with token cost + composite difficulty	2026-04-29 07:45:00 +08:00