feat(benchmark): Round 8——bridge 工具 generic typed schema + 可恢复 domain-error 契约(#100/#102) - #151
Merged
Danceiny merged 2 commits intoSep 5, 2026
Conversation
…-error 契约(issue #100/#102)——gotry_benchmark_environment 参数面 query blob → typed 三字段(action 枚举 tools/call/errors + tool + arguments additionalProperties:true,模型可见逐字段契约,与产品面 D-30 同刀法);新增 BRIDGE_ERROR_CONTRACT 封闭词表(10 码 × recoverable/remedy)经 action=errors 可拉取——模型失败后可机器取回恢复指引,不再吞进 opaque 的 child_runtime_error;失败返回的扁平形状保持不变(benchmark 诊断面依赖);tests:blob 调用点平铺改写 + errors 契约表断言 + 枚举外 action/非对象 arguments 的宿主权 ToolArgsError 断言;bridge tests 绿;tsc 绿
… into issue/100-102-kernel-round8
Base automatically changed from
issue/112-canary
to
feat/session-hotel-train-doctor
September 5, 2026 05:42
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
摘要(Round 8,承 Round 7「下一轮聚焦」)
gotry_benchmark_environment参数面 query blob → typed 三字段(action枚举 tools/call/errors required +tool+argumentsadditionalProperties:true)——模型可见逐字段契约,与产品面 D-30 同刀法;benchmark child 内普通模型对参数形状不再靠猜。BRIDGE_ERROR_CONTRACT封闭词表(10 码 × recoverable/remedy)经action=errors可拉取——模型失败后机器取回恢复指引,不再吞进 opaque 的 child_runtime_error;失败返回的扁平形状保持不变(benchmark 诊断面依赖不受扰)。验证
./scripts/run-all-tests.shALL SUITES GREEN边界
frozen treatment 的可评分终态仍需 founder 侧环境跑批;本 PR 不声称 external closure。issues #100/#102 保持 open(kernel 轨持续推进)。