Commit History

Autor SHA1 Mensaxe Data
  zhenyu.hu f2ebc9932a docs: plan benchmark ledger identity fixes hai 2 semanas
  zhenyu.hu 217db98974 fix: enforce benchmark visible ttft and call counts hai 2 semanas
  zhenyu.hu 58ede87cb6 docs: plan benchmark visible ttft call count fixes hai 2 semanas
  zhenyu.hu 11210a94bc feat: run isolated agent benchmarks hai 2 semanas
  zhenyu.hu 9464c41c94 docs: close benchmark configuration follow-ups hai 2 semanas
  zhenyu.hu f7fd343f74 refactor: unify benchmark case and mode ids hai 2 semanas
  zhenyu.hu 9ed4d56ceb docs: plan benchmark case mode single source hai 2 semanas
  zhenyu.hu a5569f70bc fix: validate benchmark urls with httpx hai 2 semanas
  zhenyu.hu 5ceda5ddc1 docs: plan benchmark url transport validation hai 2 semanas
  zhenyu.hu bd36806189 fix: reject benchmark url control characters hai 2 semanas
  zhenyu.hu d903ccc054 docs: plan strict benchmark url validation hai 2 semanas
  zhenyu.hu 5735985a46 feat: add benchmark case catalog hai 2 semanas
  zhenyu.hu ebebc5d6a7 docs: design agent benchmark runner hai 2 semanas
  zhenyu.hu c13599b4b9 docs: close invocation comparison experiment hai 2 semanas
  zhenyu.hu 368d918f8e test: enforce first reply and protocol parity hai 2 semanas
  zhenyu.hu b8476ae05d docs: plan first reply protocol matrix fixes hai 2 semanas
  zhenyu.hu 97a98a9e8b test: add invocation mode comparison matrix hai 2 semanas
  zhenyu.hu a51b2a7f21 docs: document invocation comparison experiment hai 2 semanas
  zhenyu.hu 0c62819b9c docs: plan invocation comparison closeout hai 2 semanas
  zhenyu.hu 3f441335c4 docs: close comparable metrics todos hai 2 semanas
  zhenyu.hu 1d5f65f582 fix: deduplicate metric keys during migration hai 2 semanas
  zhenyu.hu 44256c611f docs: plan metric key migration dedup hai 2 semanas
  zhenyu.hu 38d674ef48 feat: add comparable invocation metrics hai 2 semanas
  zhenyu.hu 9c15337c6b docs: plan comparable invocation metrics hai 2 semanas
  zhenyu.hu 250387d468 docs: close tool mode session restore todos hai 2 semanas
  zhenyu.hu ac526139d2 fix: reject stale replay before token advance hai 2 semanas
  zhenyu.hu 275dcae097 docs: plan stale replay token fix hai 2 semanas
  zhenyu.hu 792dcb5b29 fix: guard session replay restoration hai 2 semanas
  zhenyu.hu 9894aab5a2 docs: plan session restore race fixes hai 2 semanas
  zhenyu.hu 342a96c520 fix: make session transcript restoration atomic hai 2 semanas