docs: seal original Purge comparison and preserve network unknowns
This commit is contained in:
@@ -1,5 +1,7 @@
|
||||
# Agent Workbench项目执行规范
|
||||
|
||||
AG已一次执行并封存:正式1.25.8/a6e31dec/PID71628不变,11请求109888报告tokens+3unknown/44928本地估算。ab-off/ab-on/ba-on均connect ETIMEDOUT无响应,ba-off8请求完整matched、自主三源码/原9测试/build,自动父归集/限定Git c792069c/finalize与owner/index全过。两个开启组原版入口/apply/335模块/两个原生模块已加载,四进程树均清理;四例不等成功,Purge收益仍unknown。原50841行/119导出/工件/旧尝试/凭据行保持,新增8自有父子项目/会话,DeepSeek17→19/config118→120、原池/其余配置恢复、空闲/窗口0/9229关闭;430份47723074字节封存见docs/QA-LIVE-PURGE-AG.md。事后0HTTP的DNS/三TLS全通只证明该时点,不改写网络失败。AF0请求身份拒绝和71份封存保持;新驱动以正式API身份解决Node路径摘要差异。下一独立修Python解释器选择与终端Git/结束报告衔接,三系统入站契约反馈见docs/THREE-SYSTEM-INTERFACE-FEEDBACK-2026-11-06.md。不续跑AG/AF/旧样本/用户项目,不委派/推送/触Key,不恢复默认额度;并行Orven品牌/UI及其验证器改动不覆盖、不代交或构建。下方旧接续按历史时点。
|
||||
|
||||
AE已封存:新React/Python各10请求、合计367365报告tokens/0unknown,原九测试+tsc/build+实际1280/390页面与Python原十pytest+compileall、仅允许源码Git和owner/index均通过。原Python内置检查误选无pytest的Hermes解释器而失败,模型自主用指定解释器终端验证成功;两例未登记delivery_job/Python报告in_progress继续单列。原48100行/116导出/全部工件与旧模型尝试保持,新增2项目2会话均自有,DeepSeek11→15/config112→116、原池/其余配置/凭据状态恢复,空闲/窗口0/9229关闭。225份52216677字节封存见docs/QA-LIVE-ENGINEERING-AE.md。下一验证当前开放worker/engine契约,再另冻原版Purge AB/BA;未执行前收益unknown。不续跑AD/AE/旧样本或用户项目,不委派/推送/触Key,不恢复默认额度。下方旧接续按历史时点。
|
||||
|
||||
AD已封存:正式1.25.8/PID71628,五线10真实预检8通过;jojocode目录ECONNRESET,Claude目录UND_ERR_SOCKET及两次HTTP400缺Messages必填max_tokens原错保留。新DeepSeek Node11请求180554报告tokens、0unknown、原9测试/build/仅三模块Git88bb25cb及owner/index全过,0人工审批/继续,原高性能池已恢复。全批21请求199250报告tokens+2unknown/338本地估算,不补造计费;216份60197251字节封存、原46111行/105导出/工件精确对账,配置仅声明修订/连接编辑时间与顺序变化,9229关闭。见docs/QA-LIVE-VALIDATION-AD.md及live-validation-ad-evidence.json。7次不同文件读取、30工具schema稳定、上游122624缓存输入tokens仅本例观测;内置delivery_report仍in_progress(终端Git未登记delivery_job)如实保留。下一另冻AE React/Python,过后才原版Purge,不续跑AD/旧样本或用户项目,不委派/推送/触Key、不恢复默认额度。
|
||||
|
||||
@@ -1,5 +1,7 @@
|
||||
# AG:正式宿主身份下的原版 Purge AB/BA
|
||||
|
||||
执行状态:AG四组已一次执行并封存,11请求/109888报告tokens+3unknown;前三组连接超时,ba-off完整交付。原状态审计与430份原件封存通过,收益仍unknown。实际结果见[AG验收](QA-LIVE-PURGE-AG.md);本冻结契约和原样本不续跑。
|
||||
|
||||
AG是新批次,不续跑AF。前置AD Node、AE React/Python真实自主交付与独立核验已通过;AF在规划阶段0真实请求/0worker终止,原失败、审计和71份证据保持。本轮不改产品、不打包或升级正式1.25.8/a6e31dec/PID71628,不运行用户项目。
|
||||
|
||||
唯一实验准备变更是检测权威:冻结和逐组复核使用正式宿主的controlled-availability与purge/source API,不使用源码Node进程推断正式引擎身份。正式controlled SHA为571e9c8cfa30dca12ea5a914cf74a170c91a3bae9975c428fcafb72c8afa662a;Purge源b71679347cac069667d6cc63cb3b7736e69508eda58d50ee72746639bfc40fb0、运行输入e01d158017357cb87e233449f00828b771dad518a4118c420b6865ce15723de1。Node二进制SHA与源码环境相同,绝对路径不同仍必须按实际宿主身份核对,不放宽来源校验。
|
||||
|
||||
+1
-1
@@ -1,6 +1,6 @@
|
||||
# 1.25 计划:任务要求与交付复核
|
||||
|
||||
AF在正式协作规划时因源码Node/正式Node路径参与身份摘要而被正确拒绝,0真实请求/0worker,原状态审计及71份原件封存保持。新AG以正式检测API为权威,完整隔离四组16自有HTTP、原九测试/父Git和恢复审计通过;下一本地提交、唯一冻结AG后执行原版Purge ABBA。产品未改、Purge收益仍unknown。见[AF结果](QA-LIVE-PURGE-AF.md)与[AG计划](LIVE-PURGE-AG-PLAN.md)。下方为历史接续。
|
||||
AG原版Purge ABBA已一次执行并封存:11请求109888报告tokens+3unknown,前三组connect超时、ba-off8请求自主三源码/原9测试/build并自动父Git c792069c/finalize通过。原版两开启组已加载/apply/两个原生模块,四进程树清理;四例不等成功,收益仍unknown。原50841行/119导出/工件/旧尝试/凭据保持、原池与配置恢复,430份原件封存。AF0请求身份拒绝与71份封存保持,正式API身份已在AG解决路径摘要差异。下一独立修Python解释器选择和终端Git/结束报告登记,三系统先只读证据对接;D–F不据此完成。见[AG结果](QA-LIVE-PURGE-AG.md)与[三系统反馈](THREE-SYSTEM-INTERFACE-FEEDBACK-2026-11-06.md)。下方为历史接续。
|
||||
|
||||
AE新React/Python真实交付均已通过:20请求367365报告tokens/0unknown,原九测试/页面与原十pytest/编译、源码限定Git及owner/index独立通过,225份原证据封存;原48100行/116导出/工件保持。内置Python选错环境的失败与终端Git未登记内部交付账本保留,见[AE验收](QA-LIVE-ENGINEERING-AE.md)。下一先核对开放worker/engine契约,再另冻原版Purge AB/BA,收益仍unknown;不改变任何历史样本结论。
|
||||
|
||||
|
||||
@@ -1,5 +1,7 @@
|
||||
# AG 原版Purge对照:验证与回顾
|
||||
|
||||
AG四组已执行一次并封存:11次真实请求中8次完整matched、109888报告tokens,3次连接阶段ETIMEDOUT/无上游响应,用量unknown、合计44928本地估算不作实际消费。仅ba-off完成工程及父交付;未满足四例等成功条件,Purge工程收益仍unknown。以下准备记录按原时点保留。
|
||||
|
||||
计划见[LIVE-PURGE-AG-PLAN.md](LIVE-PURGE-AG-PLAN.md)。新批次以正式宿主检测API为身份权威,AF的0请求规划失败保持,不续跑。当前进行隔离完整驱动验证,真实AG尚未冻结或发请求,Purge工程收益仍unknown。
|
||||
|
||||
首ag-driver-cca2392e因同块batch变量遮蔽触发TDZ,在启动宿主前失败;已改为batchResult,原结果保留。封存器另修fixture结果对象被误当隔离布尔值的问题,AF原manifest不改,仅补逐SHA核对后的公开索引。后续实际结果另附,不能用夹具响应证明真实推理或收益。
|
||||
@@ -7,3 +9,36 @@
|
||||
完整隔离ag-driver-c5ff7602-52a8-440c-a24c-329d4c3257df通过:以正式Node绝对路径启动新端口/新数据宿主,实际检测路径与源码不同但二进制SHA相同;正式API权威贯穿四组,16自有HTTP、每组原九测试/build、父工程限定Git/owner/index、原版入口/apply/两个原生模块及进程清理均通过。原11行/1导出/工件/凭据状态保持,新增8父子项目/8会话/16尝试全部自有,最终审计9c83415c通过;477份19575831字节封存且逐SHA核对。真实模型请求0。
|
||||
|
||||
回顾:AF原因在不同Node绝对路径参与身份摘要;本次夹具复现同样环境差异并以正式检测API通过,不放宽产品核对。有效驱动语法和diff检查通过,AF公开索引离线恢复通过,正式程序未变。准备本地提交后唯一冻结AG,再运行一次四组真实对照。
|
||||
|
||||
## 真实四组结果
|
||||
|
||||
冻结驱动a6830bbf,正式1.25.8/a6e31dec/PID71628;原94项目119会话、DeepSeek17/config118。来源由正式API核对,四组同初始Git内容、同工具schema,逐真实HTTP均无输出cap、模型历史full/inputBudget=null。没有实现提示、人工批准、继续、重试或换线。
|
||||
|
||||
| 组 | 真实请求 | 实际报告用量 | 结果 | 总墙钟 |
|
||||
|---|---:|---|---|---:|
|
||||
| ab-off | 1 | unknown,本地估算9024 | connect ETIMEDOUT,无响应/工具/修改;原9项仍失败 | 37881ms |
|
||||
| ab-on | 1 | unknown,本地估算17952 | connect ETIMEDOUT,无响应/工具/修改;原9项仍失败 | 40877ms |
|
||||
| ba-on | 1 | unknown,本地估算17952 | connect ETIMEDOUT,无响应/工具/修改;原9项仍失败 | 42150ms |
|
||||
| ba-off | 8 | 104720输入+5168输出=109888 | 自主三源码、原9/9及build、自动父归集/检查/Git/finalize通过 | 112608ms |
|
||||
|
||||
前三次诊断均network.stage=connect、responseReceived=false、usageStatus=unknown、identity=incomplete;没有上游型号可以核对,不是型号mismatch,也不能将零报告tokens当零消费。原POST不重发。第三组之后连接自然恢复,第四组全部matched;不因此宣称稳定或已修复网络。
|
||||
|
||||
ba-off真实模型只修改src/amount.mjs、src/summary.mjs、src/report.mjs,使用plan_checks/run_checks取得当前test/build通过后结束;父流程自动归集、合并、再次检查、限定源码提交c792069c571a61dcf563d2bebf60cabcfe8d581a并finalize。独立最终原九测试9/9/build通过,当前源码=提交,owner原字节/非目标暂存与未暂存/未跟踪、子HEAD/index保持。自动父动作和模型子工程行为分开,失败三组没有实际父归集或提交。9次读取各为不同文件,shadow重复读0仅为本例;上游报告90624缓存输入tokens,不作普遍不掉缓存或Purge/C因果收益。
|
||||
|
||||
两个开启组在网络失败前确实加载原版入口并调用apply,335个模块与两个原生模块有记录,60源文件/393运行输入身份一致。四组原生根进程退出、所属树清理均核对。开启组engine_run失败由连接错误终结,不能把未完成工程描述为插件兼容或收益已通过;加载证据与工程完成状态分开。
|
||||
|
||||
真实11份请求的工具SHA一致为e2d96a46ea387a7583272d2a0c5aaa339713aeb83599351d4dd37e510c191efe;开启首请求的本地估算大于关闭组不代表实际账单或压缩收益。首token延迟和部分工具时长没有原账本字段,保留null,不编造。四例不等成功,comparison.eligible=false,不计算节省率。
|
||||
|
||||
## 原状态、网络观察与封存
|
||||
|
||||
最终审计final-36ea1aba通过:原50841行、119导出、全部旧工件、模型尝试和凭据行精确保持;新增8父子项目/8会话、978事件及outbox、335记录、207去重工件全部自有。DeepSeek17→19、config118→120仅声明的暂选/恢复修订和编辑时间,原高性能池/其余配置/credentialRevision1/五条Key存在状态恢复。正式仍PID71628/原instance、空闲/窗口0/9229关闭,未升级、重启或触真实Key。
|
||||
|
||||
430份47723074字节原件封存并逐SHA通过,见[live-purge-ag-evidence.json](live-purge-ag-evidence.json)和.qa/live-purge-ag/FINAL-MANIFEST.json;AF及旧样本原件不重打包。保存全部失败、原请求/响应、engine输入、原检查/Git、配置恢复与原状态审计。
|
||||
|
||||
封存后另用正式Node路径启动全新只读观察进程,0 HTTP/模型请求、无Key/配置操作。系统DNS当时返回123.125.246.121和119.188.175.46(TTL19),主机名和两地址TLSv1.3证书校验均通过,分别43/28/181ms,进程代理环境三项false。见[live-purge-ag-network-evidence.json](live-purge-ag-network-evidence.json)。该观察不复原失败时的地址选择/宿主环境,也不证明持续可达;不改DNS、代理、超时或重发POST,原AG封存不变。
|
||||
|
||||
## 回顾与下一切片
|
||||
|
||||
正式宿主身份问题已由完整隔离夹具及真实四组规划/启动修复验证;原AF失败保持。本轮实证了开放原生Harness的一个完整自主工程与自动父交付,但原版Purge两组都在收到模型输出前网络失败,工程收益尚未验证。目录网络、Claude必填max_tokens边界、历史真实预留/重复读取/缓存未知均保留。
|
||||
|
||||
下一产品切片分别处理Python检查解释器显式选择/不可变绑定,以及终端Git与内部报告登记/运行结束状态衔接;保留AD/AE旧失败,不用重放旧样本补签。三系统入站协议另有[接口反馈](THREE-SYSTEM-INTERFACE-FEEDBACK-2026-11-06.md)。并行Orven品牌/UI源码不覆盖、不代交或构建,不用本轮验证替它验收。只本地提交,不委派、不推送。
|
||||
|
||||
@@ -25,3 +25,5 @@ Python内置plan_checks/run_checks在AE选中无pytest的Hermes解释器,原
|
||||
先离线只读消费一份已完成工程的脱敏回执:来源版本、请求摘要、项目/Session/Task/Run、检查/页面/Git工件ID及SHA全部可核对,导入认知侧为候选,0模型请求/0工程写入。先定义engineering-delegation/v1草案(这是建议的新契约,尚未实现),再用全新合成工程和自有固定HTTP验证入站启动、原游标查询和停止。
|
||||
|
||||
验收覆盖同key同参返回同任务、同key异参拒绝、响应丢失先查询不重发、重启恢复、外部源码漂移、失败/unknown、取消接收与所属进程确已退出分开、已有Git但内部报告缺失。一次委托只产生一次允许写入/一次提交,实际测试/页面与当前commit绑定。委托若要求额外范围或费用条件而通道不能兑现,应在启动前说明;不修改用户全局权限或截断历史来伪称已满足。以上通过后再另冻真实调用与质量样本,三方原质量失败和Purge unknown不因接口通过而关闭。
|
||||
|
||||
阶段补充:本反馈后AG已一次执行,11真实请求中8次完整matched/109888报告tokens、3次connect ETIMEDOUT用量unknown;仅关闭组ba-off自主原九测试/build通过,由自动父流程限定Git/finalize交付。两开启组原版入口/apply/两个原生模块加载与进程清理均有证据,但无模型响应,Purge工程收益仍unknown。430份证据及原50841行/119导出/工件/尝试/凭据对账见[AG结果](QA-LIVE-PURGE-AG.md)。这不替代拟议入站协议的真实验收,Python/终端Git缺口仍按上文处理。
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Tianshu 研究后的 Agent Workbench 采用计划
|
||||
|
||||
AF在正式协作规划时因源码Node/正式Node路径参与身份摘要而被正确拒绝,0真实请求/0worker,原状态审计及71份原件封存保持。新AG以正式检测API为权威,完整隔离四组16自有HTTP、原九测试/父Git和恢复审计通过;下一本地提交、唯一冻结AG后执行原版Purge ABBA。产品未改、Purge收益仍unknown。见[AF结果](QA-LIVE-PURGE-AF.md)与[AG计划](LIVE-PURGE-AG-PLAN.md)。下方为历史接续。
|
||||
AG原版Purge ABBA已一次执行并封存:11请求109888报告tokens+3unknown,前三组connect超时、ba-off8请求自主三源码/原9测试/build并自动父Git c792069c/finalize通过。原版两开启组已加载/apply/两个原生模块,四进程树清理;四例不等成功,收益仍unknown。原50841行/119导出/工件/旧尝试/凭据保持、原池与配置恢复,430份原件封存。AF0请求身份拒绝与71份封存保持,正式API身份已在AG解决路径摘要差异。下一独立修Python解释器选择和终端Git/结束报告登记,三系统先只读证据对接;D–F不据此完成。见[AG结果](QA-LIVE-PURGE-AG.md)与[三系统反馈](THREE-SYSTEM-INTERFACE-FEEDBACK-2026-11-06.md)。下方为历史接续。
|
||||
|
||||
AE新React/Python真实交付均已通过:20请求367365报告tokens/0unknown,原九测试/页面与原十pytest/编译、源码限定Git及owner/index独立通过,225份原证据封存;原48100行/116导出/工件保持。内置Python选错环境的失败与终端Git未登记内部交付账本保留,见[AE验收](QA-LIVE-ENGINEERING-AE.md)。下一先核对开放worker/engine契约,再另冻原版Purge AB/BA,收益仍unknown;不改变任何历史样本结论。
|
||||
|
||||
|
||||
@@ -0,0 +1,615 @@
|
||||
{
|
||||
"summary": {
|
||||
"schema": 1,
|
||||
"version": "1.25.8",
|
||||
"driverCommit": "a6830bbf501d6871bee21b0ccc7e613505078fc3",
|
||||
"samples": [
|
||||
{
|
||||
"sample": "ab-off",
|
||||
"enabled": false,
|
||||
"success": false,
|
||||
"failure": {
|
||||
"reason": "model_or_engine_stopped",
|
||||
"groupStatus": "paused",
|
||||
"summary": "子任务已停止或需要处理,已完成结果保留。"
|
||||
},
|
||||
"final": {
|
||||
"test": {
|
||||
"status": 1,
|
||||
"signal": null,
|
||||
"error": null,
|
||||
"stats": {
|
||||
"tests": 9,
|
||||
"pass": 0,
|
||||
"fail": 9,
|
||||
"cancelled": 0,
|
||||
"skipped": 0,
|
||||
"todo": 0
|
||||
},
|
||||
"stdoutSha256": "3131a52f24b58ef36299e3e5fa892e2fc24a4707a77a13a380207e2b085e0ebf",
|
||||
"stderrSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
|
||||
},
|
||||
"build": {
|
||||
"status": 0,
|
||||
"signal": null,
|
||||
"error": null,
|
||||
"stats": null,
|
||||
"stdoutSha256": "9307ee4e4f72392dd2f29f26c8e9fd3c8e55353bbe21f93627f260f108be9c2c",
|
||||
"stderrSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
|
||||
}
|
||||
},
|
||||
"commit": "c213ad68236af508bf9b31e9739483caeda4ef0c",
|
||||
"initialHead": "c213ad68236af508bf9b31e9739483caeda4ef0c",
|
||||
"committedPaths": [],
|
||||
"committedCurrent": true,
|
||||
"parentSessionId": "03fd7d82-5e71-46d9-a191-7842c137276e",
|
||||
"workerSessionId": "4064d1fa-4bf2-4f26-949a-e20e34c80463",
|
||||
"ownerBytesAndIndexPreserved": true,
|
||||
"supervisorApprovals": 0,
|
||||
"supervisorContinuations": 0,
|
||||
"automaticParentCollectionAndChecks": true,
|
||||
"identitiesPassed": false,
|
||||
"nativePassed": false,
|
||||
"requests": 1,
|
||||
"usage": {
|
||||
"requests": 1,
|
||||
"inputTokens": 0,
|
||||
"outputTokens": 0,
|
||||
"totalTokens": 0,
|
||||
"unknownRequests": 1,
|
||||
"estimatedUnknownTokens": 9024
|
||||
},
|
||||
"reportedCachedInputTokens": 0,
|
||||
"elapsedMs": 37881,
|
||||
"engineElapsedMs": 24149,
|
||||
"modelIdentityPolicy": "require-match",
|
||||
"reads": [],
|
||||
"requestTimings": [
|
||||
{
|
||||
"id": "7917bf76-3293-41c1-be2e-63bc71fa6402",
|
||||
"startedAt": "2026-11-06T05:54:35.949Z",
|
||||
"finishedAt": "2026-11-06T05:54:57.298Z",
|
||||
"elapsedMs": 21349,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
}
|
||||
],
|
||||
"toolCalls": [],
|
||||
"shadowRepeatedReads": 0,
|
||||
"originalPurge": [
|
||||
{
|
||||
"id": "8ec6e323-c963-44b3-998f-d02bb5d7509b",
|
||||
"kind": null,
|
||||
"sourceSha256": null,
|
||||
"entryLoaded": false,
|
||||
"applyCalled": false,
|
||||
"loadedModules": 0,
|
||||
"nativeLoads": 0,
|
||||
"rootExited": true,
|
||||
"treeConfirmed": true
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"sample": "ab-on",
|
||||
"enabled": true,
|
||||
"success": false,
|
||||
"failure": {
|
||||
"reason": "model_or_engine_stopped",
|
||||
"groupStatus": "paused",
|
||||
"summary": "子任务已停止或需要处理,已完成结果保留。"
|
||||
},
|
||||
"final": {
|
||||
"test": {
|
||||
"status": 1,
|
||||
"signal": null,
|
||||
"error": null,
|
||||
"stats": {
|
||||
"tests": 9,
|
||||
"pass": 0,
|
||||
"fail": 9,
|
||||
"cancelled": 0,
|
||||
"skipped": 0,
|
||||
"todo": 0
|
||||
},
|
||||
"stdoutSha256": "ab0a04380f1efd6b6789d5827b0c1ecad43749bf73f91457d0a2099a5fbe4da9",
|
||||
"stderrSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
|
||||
},
|
||||
"build": {
|
||||
"status": 0,
|
||||
"signal": null,
|
||||
"error": null,
|
||||
"stats": null,
|
||||
"stdoutSha256": "9307ee4e4f72392dd2f29f26c8e9fd3c8e55353bbe21f93627f260f108be9c2c",
|
||||
"stderrSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
|
||||
}
|
||||
},
|
||||
"commit": "c213ad68236af508bf9b31e9739483caeda4ef0c",
|
||||
"initialHead": "c213ad68236af508bf9b31e9739483caeda4ef0c",
|
||||
"committedPaths": [],
|
||||
"committedCurrent": true,
|
||||
"parentSessionId": "90fa6590-7405-455c-a5f2-a3e507593b16",
|
||||
"workerSessionId": "fdce00e6-0716-4cdc-b086-98be8712746f",
|
||||
"ownerBytesAndIndexPreserved": true,
|
||||
"supervisorApprovals": 0,
|
||||
"supervisorContinuations": 0,
|
||||
"automaticParentCollectionAndChecks": true,
|
||||
"identitiesPassed": false,
|
||||
"nativePassed": false,
|
||||
"requests": 1,
|
||||
"usage": {
|
||||
"requests": 1,
|
||||
"inputTokens": 0,
|
||||
"outputTokens": 0,
|
||||
"totalTokens": 0,
|
||||
"unknownRequests": 1,
|
||||
"estimatedUnknownTokens": 17952
|
||||
},
|
||||
"reportedCachedInputTokens": 0,
|
||||
"elapsedMs": 40877,
|
||||
"engineElapsedMs": 28088,
|
||||
"modelIdentityPolicy": "require-match",
|
||||
"reads": [],
|
||||
"requestTimings": [
|
||||
{
|
||||
"id": "6b985ab2-0f73-413c-b05d-ef8c49184a42",
|
||||
"startedAt": "2026-11-06T05:55:16.836Z",
|
||||
"finishedAt": "2026-11-06T05:55:38.216Z",
|
||||
"elapsedMs": 21380,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
}
|
||||
],
|
||||
"toolCalls": [],
|
||||
"shadowRepeatedReads": 0,
|
||||
"originalPurge": [
|
||||
{
|
||||
"id": "fb8c79d5-2c84-4f3e-940a-e0eb1f54be6c",
|
||||
"kind": "purge-original-restricted-v1",
|
||||
"sourceSha256": "b71679347cac069667d6cc63cb3b7736e69508eda58d50ee72746639bfc40fb0",
|
||||
"entryLoaded": true,
|
||||
"applyCalled": true,
|
||||
"loadedModules": 335,
|
||||
"nativeLoads": 2,
|
||||
"rootExited": true,
|
||||
"treeConfirmed": true
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"sample": "ba-on",
|
||||
"enabled": true,
|
||||
"success": false,
|
||||
"failure": {
|
||||
"reason": "model_or_engine_stopped",
|
||||
"groupStatus": "paused",
|
||||
"summary": "子任务已停止或需要处理,已完成结果保留。"
|
||||
},
|
||||
"final": {
|
||||
"test": {
|
||||
"status": 1,
|
||||
"signal": null,
|
||||
"error": null,
|
||||
"stats": {
|
||||
"tests": 9,
|
||||
"pass": 0,
|
||||
"fail": 9,
|
||||
"cancelled": 0,
|
||||
"skipped": 0,
|
||||
"todo": 0
|
||||
},
|
||||
"stdoutSha256": "78011e22cd774f0055c62e73caa437de2478bd30f8b8cd334fb661b65be6364b",
|
||||
"stderrSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
|
||||
},
|
||||
"build": {
|
||||
"status": 0,
|
||||
"signal": null,
|
||||
"error": null,
|
||||
"stats": null,
|
||||
"stdoutSha256": "9307ee4e4f72392dd2f29f26c8e9fd3c8e55353bbe21f93627f260f108be9c2c",
|
||||
"stderrSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
|
||||
}
|
||||
},
|
||||
"commit": "c213ad68236af508bf9b31e9739483caeda4ef0c",
|
||||
"initialHead": "c213ad68236af508bf9b31e9739483caeda4ef0c",
|
||||
"committedPaths": [],
|
||||
"committedCurrent": true,
|
||||
"parentSessionId": "8dc55f73-a245-4b05-ba65-c347118c16ca",
|
||||
"workerSessionId": "27fd4b03-f035-423b-8937-b6a6b2f478a7",
|
||||
"ownerBytesAndIndexPreserved": true,
|
||||
"supervisorApprovals": 0,
|
||||
"supervisorContinuations": 0,
|
||||
"automaticParentCollectionAndChecks": true,
|
||||
"identitiesPassed": false,
|
||||
"nativePassed": false,
|
||||
"requests": 1,
|
||||
"usage": {
|
||||
"requests": 1,
|
||||
"inputTokens": 0,
|
||||
"outputTokens": 0,
|
||||
"totalTokens": 0,
|
||||
"unknownRequests": 1,
|
||||
"estimatedUnknownTokens": 17952
|
||||
},
|
||||
"reportedCachedInputTokens": 0,
|
||||
"elapsedMs": 42150,
|
||||
"engineElapsedMs": 27983,
|
||||
"modelIdentityPolicy": "require-match",
|
||||
"reads": [],
|
||||
"requestTimings": [
|
||||
{
|
||||
"id": "0d1c9abe-a59a-4ae9-8ad5-bbf892582e17",
|
||||
"startedAt": "2026-11-06T05:55:59.200Z",
|
||||
"finishedAt": "2026-11-06T05:56:20.575Z",
|
||||
"elapsedMs": 21375,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
}
|
||||
],
|
||||
"toolCalls": [],
|
||||
"shadowRepeatedReads": 0,
|
||||
"originalPurge": [
|
||||
{
|
||||
"id": "1fc7055e-2481-4f49-adb3-91c67c983ede",
|
||||
"kind": "purge-original-restricted-v1",
|
||||
"sourceSha256": "b71679347cac069667d6cc63cb3b7736e69508eda58d50ee72746639bfc40fb0",
|
||||
"entryLoaded": true,
|
||||
"applyCalled": true,
|
||||
"loadedModules": 335,
|
||||
"nativeLoads": 2,
|
||||
"rootExited": true,
|
||||
"treeConfirmed": true
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"sample": "ba-off",
|
||||
"enabled": false,
|
||||
"success": true,
|
||||
"failure": null,
|
||||
"final": {
|
||||
"test": {
|
||||
"status": 0,
|
||||
"signal": null,
|
||||
"error": null,
|
||||
"stats": {
|
||||
"tests": 9,
|
||||
"pass": 9,
|
||||
"fail": 0,
|
||||
"cancelled": 0,
|
||||
"skipped": 0,
|
||||
"todo": 0
|
||||
},
|
||||
"stdoutSha256": "d9fa18b4c938d61a00d0b5e9274a00ceba9cc4b6fb1a4a735eea8c44b3ae4fb0",
|
||||
"stderrSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
|
||||
},
|
||||
"build": {
|
||||
"status": 0,
|
||||
"signal": null,
|
||||
"error": null,
|
||||
"stats": null,
|
||||
"stdoutSha256": "9307ee4e4f72392dd2f29f26c8e9fd3c8e55353bbe21f93627f260f108be9c2c",
|
||||
"stderrSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
|
||||
}
|
||||
},
|
||||
"commit": "c792069c571a61dcf563d2bebf60cabcfe8d581a",
|
||||
"initialHead": "c213ad68236af508bf9b31e9739483caeda4ef0c",
|
||||
"committedPaths": [
|
||||
"src/amount.mjs",
|
||||
"src/report.mjs",
|
||||
"src/summary.mjs"
|
||||
],
|
||||
"committedCurrent": true,
|
||||
"parentSessionId": "618e2818-2c8f-48f1-897d-be6ee48c1406",
|
||||
"workerSessionId": "91d910d3-7849-4f43-a0d5-1cfa019c34d7",
|
||||
"ownerBytesAndIndexPreserved": true,
|
||||
"supervisorApprovals": 0,
|
||||
"supervisorContinuations": 0,
|
||||
"automaticParentCollectionAndChecks": true,
|
||||
"identitiesPassed": true,
|
||||
"nativePassed": true,
|
||||
"requests": 8,
|
||||
"usage": {
|
||||
"requests": 8,
|
||||
"inputTokens": 104720,
|
||||
"outputTokens": 5168,
|
||||
"totalTokens": 109888,
|
||||
"unknownRequests": 0,
|
||||
"estimatedUnknownTokens": 0
|
||||
},
|
||||
"reportedCachedInputTokens": 90624,
|
||||
"elapsedMs": 112608,
|
||||
"engineElapsedMs": 63665,
|
||||
"modelIdentityPolicy": "require-match",
|
||||
"reads": [
|
||||
{
|
||||
"path": "src/amount.mjs"
|
||||
},
|
||||
{
|
||||
"path": "src/summary.mjs"
|
||||
},
|
||||
{
|
||||
"path": "src/report.mjs"
|
||||
},
|
||||
{
|
||||
"path": "test/expense.test.mjs"
|
||||
},
|
||||
{
|
||||
"path": "package.json"
|
||||
},
|
||||
{
|
||||
"path": "scripts/build.mjs"
|
||||
},
|
||||
{
|
||||
"path": "README.md"
|
||||
},
|
||||
{
|
||||
"path": "owner-staged.txt"
|
||||
},
|
||||
{
|
||||
"path": "owner-unstaged.txt"
|
||||
}
|
||||
],
|
||||
"requestTimings": [
|
||||
{
|
||||
"id": "189f61cd-e960-4eec-bb13-8e637447c40b",
|
||||
"startedAt": "2026-11-06T05:56:38.487Z",
|
||||
"finishedAt": "2026-11-06T05:56:40.929Z",
|
||||
"elapsedMs": 2442,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
},
|
||||
{
|
||||
"id": "4a18ae73-0023-41fe-9754-f792501f5d8f",
|
||||
"startedAt": "2026-11-06T05:56:48.178Z",
|
||||
"finishedAt": "2026-11-06T05:56:49.703Z",
|
||||
"elapsedMs": 1525,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
},
|
||||
{
|
||||
"id": "9f75709b-f0f0-4b6e-b206-b0128b40aaf5",
|
||||
"startedAt": "2026-11-06T05:56:55.321Z",
|
||||
"finishedAt": "2026-11-06T05:57:04.779Z",
|
||||
"elapsedMs": 9458,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
},
|
||||
{
|
||||
"id": "76f12dd5-abd9-4e63-b8bc-1359f971cc49",
|
||||
"startedAt": "2026-11-06T05:57:06.534Z",
|
||||
"finishedAt": "2026-11-06T05:57:18.859Z",
|
||||
"elapsedMs": 12325,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
},
|
||||
{
|
||||
"id": "d12b6f18-4020-4d93-8d3d-bec8d2d77197",
|
||||
"startedAt": "2026-11-06T05:57:22.601Z",
|
||||
"finishedAt": "2026-11-06T05:57:24.162Z",
|
||||
"elapsedMs": 1561,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
},
|
||||
{
|
||||
"id": "62853fe6-40a2-4f45-a4c0-8977f7cb5f46",
|
||||
"startedAt": "2026-11-06T05:57:25.651Z",
|
||||
"finishedAt": "2026-11-06T05:57:26.677Z",
|
||||
"elapsedMs": 1026,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
},
|
||||
{
|
||||
"id": "16cdb866-f27d-4db8-81f7-b0e81d9e8c08",
|
||||
"startedAt": "2026-11-06T05:57:28.471Z",
|
||||
"finishedAt": "2026-11-06T05:57:29.203Z",
|
||||
"elapsedMs": 732,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
},
|
||||
{
|
||||
"id": "4c133bf8-b9f3-47a8-be57-848549afbd10",
|
||||
"startedAt": "2026-11-06T05:57:35.206Z",
|
||||
"finishedAt": "2026-11-06T05:57:39.484Z",
|
||||
"elapsedMs": 4278,
|
||||
"firstTokenLatencyMs": null,
|
||||
"firstTokenLatencyNote": "The attempt ledger does not record this metric; no estimate substituted."
|
||||
}
|
||||
],
|
||||
"toolCalls": [
|
||||
{
|
||||
"tool": "inspect_project",
|
||||
"startedAt": "2026-11-06T05:56:41.209Z",
|
||||
"finishedAt": "2026-11-06T05:56:41.747Z",
|
||||
"durationMs": 538
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:42.342Z",
|
||||
"finishedAt": "2026-11-06T05:56:42.574Z",
|
||||
"durationMs": 232
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:43.385Z",
|
||||
"finishedAt": "2026-11-06T05:56:43.622Z",
|
||||
"durationMs": 237
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:44.720Z",
|
||||
"finishedAt": "2026-11-06T05:56:45.119Z",
|
||||
"durationMs": 399
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:46.013Z",
|
||||
"finishedAt": "2026-11-06T05:56:46.258Z",
|
||||
"durationMs": 245
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:47.161Z",
|
||||
"finishedAt": "2026-11-06T05:56:47.410Z",
|
||||
"durationMs": 249
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:50.041Z",
|
||||
"finishedAt": "2026-11-06T05:56:50.292Z",
|
||||
"durationMs": 251
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:51.077Z",
|
||||
"finishedAt": "2026-11-06T05:56:51.313Z",
|
||||
"durationMs": 236
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:53.255Z",
|
||||
"finishedAt": "2026-11-06T05:56:53.526Z",
|
||||
"durationMs": 271
|
||||
},
|
||||
{
|
||||
"tool": "read_file",
|
||||
"startedAt": "2026-11-06T05:56:54.252Z",
|
||||
"finishedAt": "2026-11-06T05:56:54.488Z",
|
||||
"durationMs": 236
|
||||
},
|
||||
{
|
||||
"tool": "plan_checks",
|
||||
"startedAt": "2026-11-06T05:57:05.063Z",
|
||||
"finishedAt": "2026-11-06T05:57:05.791Z",
|
||||
"durationMs": 728
|
||||
},
|
||||
{
|
||||
"tool": "propose_file",
|
||||
"startedAt": "2026-11-06T05:57:19.151Z",
|
||||
"finishedAt": "2026-11-06T05:57:19.630Z",
|
||||
"durationMs": 479
|
||||
},
|
||||
{
|
||||
"tool": "propose_file",
|
||||
"startedAt": "2026-11-06T05:57:20.368Z",
|
||||
"finishedAt": "2026-11-06T05:57:20.639Z",
|
||||
"durationMs": 271
|
||||
},
|
||||
{
|
||||
"tool": "propose_file",
|
||||
"startedAt": "2026-11-06T05:57:21.726Z",
|
||||
"finishedAt": "2026-11-06T05:57:21.993Z",
|
||||
"durationMs": 267
|
||||
},
|
||||
{
|
||||
"tool": "plan_checks",
|
||||
"startedAt": "2026-11-06T05:57:26.984Z",
|
||||
"finishedAt": "2026-11-06T05:57:27.696Z",
|
||||
"durationMs": 712
|
||||
},
|
||||
{
|
||||
"tool": "run_checks",
|
||||
"startedAt": "2026-11-06T05:57:29.517Z",
|
||||
"finishedAt": "2026-11-06T05:57:34.437Z",
|
||||
"durationMs": 4920
|
||||
}
|
||||
],
|
||||
"shadowRepeatedReads": 0,
|
||||
"originalPurge": [
|
||||
{
|
||||
"id": "d1e4bb2d-2051-4b40-b6a6-707e200aeec8",
|
||||
"kind": null,
|
||||
"sourceSha256": null,
|
||||
"entryLoaded": false,
|
||||
"applyCalled": false,
|
||||
"loadedModules": 0,
|
||||
"nativeLoads": 0,
|
||||
"rootExited": true,
|
||||
"treeConfirmed": true
|
||||
}
|
||||
]
|
||||
}
|
||||
],
|
||||
"requests": 11,
|
||||
"reportedRequests": 8,
|
||||
"reportedTokens": 109888,
|
||||
"unknownRequests": 3,
|
||||
"unknownLocalEstimate": 44928,
|
||||
"actualToolSha256": [
|
||||
"e2d96a46ea387a7583272d2a0c5aaa339713aeb83599351d4dd37e510c191efe"
|
||||
],
|
||||
"sameTools": true,
|
||||
"comparison": {
|
||||
"eligible": false,
|
||||
"reason": "Four equal-success samples with stable tools are required; early failure is not token savings."
|
||||
},
|
||||
"originalStateAudit": "audits/final-36ea1aba-0b35-4a61-94e0-95ec8b65ddd6/result.json",
|
||||
"originalStateAuditResult": {
|
||||
"success": true,
|
||||
"version": "1.25.8",
|
||||
"pid": 71628,
|
||||
"originalProjects": 94,
|
||||
"originalSessions": 119,
|
||||
"addedProjects": 8,
|
||||
"addedSessions": 8,
|
||||
"checkedOriginalRows": 50841,
|
||||
"addedRows": {
|
||||
"events": 978,
|
||||
"metadata": 0,
|
||||
"outbox": 978,
|
||||
"projects": 8,
|
||||
"records": 335,
|
||||
"sessions": 8,
|
||||
"sqlite_sequence": 0
|
||||
},
|
||||
"newArtifactCount": 207,
|
||||
"originalExports": "exact",
|
||||
"originalArtifactBytes": "exact",
|
||||
"originalCredentialRow": "exact hash; no decryption or copy",
|
||||
"allOldModelAttempts": "exact",
|
||||
"newModelAttempts": 11,
|
||||
"configComparison": {
|
||||
"original": 118,
|
||||
"current": 120,
|
||||
"allOtherFields": "exact"
|
||||
},
|
||||
"connectionChanges": [
|
||||
{
|
||||
"id": "1a7a55a9-29f2-4b9f-8ea2-5a55afd7eacd",
|
||||
"originalRevision": 17,
|
||||
"currentRevision": 19,
|
||||
"originalUpdatedAt": "2026-11-06T05:42:01.952Z",
|
||||
"currentUpdatedAt": "2026-11-06T05:58:15.837Z"
|
||||
}
|
||||
],
|
||||
"connectionOrder": {
|
||||
"before": [
|
||||
"5b4daed9-b2f7-4de0-b949-573eba70c7b3",
|
||||
"93d0cf6d-4f32-4918-9da8-6f125aa8a6e5",
|
||||
"09b0293b-faad-45c3-832d-7c8b615813aa",
|
||||
"9ae348a5-d0a6-42ed-8664-0da5f9effe34",
|
||||
"1a7a55a9-29f2-4b9f-8ea2-5a55afd7eacd"
|
||||
],
|
||||
"after": [
|
||||
"5b4daed9-b2f7-4de0-b949-573eba70c7b3",
|
||||
"93d0cf6d-4f32-4918-9da8-6f125aa8a6e5",
|
||||
"09b0293b-faad-45c3-832d-7c8b615813aa",
|
||||
"9ae348a5-d0a6-42ed-8664-0da5f9effe34",
|
||||
"1a7a55a9-29f2-4b9f-8ea2-5a55afd7eacd"
|
||||
]
|
||||
},
|
||||
"originalRows": "exact except declared project positions, sequence and validated public metadata",
|
||||
"keysReadOrResaved": false,
|
||||
"inspectorClosed": true
|
||||
},
|
||||
"keysReadOrResaved": false,
|
||||
"historicalSamplesReplayed": false,
|
||||
"priorFailedExperimentsRemainUnchanged": true
|
||||
},
|
||||
"evidence": {
|
||||
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\live-purge-ag",
|
||||
"files": 430,
|
||||
"bytes": 47723074,
|
||||
"manifestSha256": "653e88ef15561b4f3b864b5881e91b716d09c152a8e1ddd02853a9a756bfb898"
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,101 @@
|
||||
{
|
||||
"schemaVersion": 1,
|
||||
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\ag-network-8c22fecb-a982-4b84-9ff3-e00902c549d4",
|
||||
"files": [
|
||||
{
|
||||
"path": "observer.mjs",
|
||||
"bytes": 4020,
|
||||
"sha256": "3cd3b8cc34e25e25f8029c56ae1c9fe5f8c7d324c14bce65f9178eff1984f203"
|
||||
},
|
||||
{
|
||||
"path": "result.json",
|
||||
"bytes": 2527,
|
||||
"sha256": "f5b49bf6b5d9adb06b4a6cf4327543602aba3a490cd36a270157b4b88b0fbcce"
|
||||
}
|
||||
],
|
||||
"observation": {
|
||||
"schemaVersion": 1,
|
||||
"startedAt": "2026-11-06T06:00:15.497Z",
|
||||
"finishedAt": "2026-11-06T06:00:15.773Z",
|
||||
"hostname": "api.deepseek.com",
|
||||
"port": 443,
|
||||
"executable": "C:\\Users\\wangy\\AppData\\Local\\Programs\\AgentWorkbench\\versions\\Agent-Workbench-1.25.5-win-x64-20261106003510201\\runtime\\node.exe",
|
||||
"executableSha256": "d702daa815eabbff763c7d3808e235cf031cf2ca70211b56847bc72f51e240e5",
|
||||
"observerSha256": "3cd3b8cc34e25e25f8029c56ae1c9fe5f8c7d324c14bce65f9178eff1984f203",
|
||||
"dns": {
|
||||
"system": {
|
||||
"success": true,
|
||||
"values": [
|
||||
{
|
||||
"address": "123.125.246.121",
|
||||
"family": 4
|
||||
},
|
||||
{
|
||||
"address": "119.188.175.46",
|
||||
"family": 4
|
||||
}
|
||||
]
|
||||
},
|
||||
"ipv4": {
|
||||
"success": true,
|
||||
"values": [
|
||||
{
|
||||
"address": "123.125.246.121",
|
||||
"ttl": 19
|
||||
},
|
||||
{
|
||||
"address": "119.188.175.46",
|
||||
"ttl": 19
|
||||
}
|
||||
]
|
||||
},
|
||||
"ipv6": {
|
||||
"success": true,
|
||||
"values": []
|
||||
}
|
||||
},
|
||||
"probes": [
|
||||
{
|
||||
"label": "hostname",
|
||||
"host": "api.deepseek.com",
|
||||
"elapsedMs": 43,
|
||||
"success": true,
|
||||
"remoteAddress": "123.125.246.121",
|
||||
"remoteFamily": "IPv4",
|
||||
"protocol": "TLSv1.3",
|
||||
"certificateFingerprint256": "30:D8:25:29:D1:9C:5D:A6:17:5A:80:DB:1C:55:22:AC:3B:A5:B4:25:89:21:24:05:CD:C8:78:3A:1B:F0:47:FA"
|
||||
},
|
||||
{
|
||||
"label": "resolved-address",
|
||||
"host": "123.125.246.121",
|
||||
"elapsedMs": 28,
|
||||
"success": true,
|
||||
"remoteAddress": "123.125.246.121",
|
||||
"remoteFamily": "IPv4",
|
||||
"protocol": "TLSv1.3",
|
||||
"certificateFingerprint256": "30:D8:25:29:D1:9C:5D:A6:17:5A:80:DB:1C:55:22:AC:3B:A5:B4:25:89:21:24:05:CD:C8:78:3A:1B:F0:47:FA"
|
||||
},
|
||||
{
|
||||
"label": "resolved-address",
|
||||
"host": "119.188.175.46",
|
||||
"elapsedMs": 181,
|
||||
"success": true,
|
||||
"remoteAddress": "119.188.175.46",
|
||||
"remoteFamily": "IPv4",
|
||||
"protocol": "TLSv1.3",
|
||||
"certificateFingerprint256": "30:D8:25:29:D1:9C:5D:A6:17:5A:80:DB:1C:55:22:AC:3B:A5:B4:25:89:21:24:05:CD:C8:78:3A:1B:F0:47:FA"
|
||||
}
|
||||
],
|
||||
"proxyEnvironmentPresent": {
|
||||
"http": false,
|
||||
"https": false,
|
||||
"noProxy": false
|
||||
},
|
||||
"httpRequests": 0,
|
||||
"modelRequests": 0,
|
||||
"keysReadOrResaved": false,
|
||||
"productConfigurationChanged": false,
|
||||
"scope": "Fresh process DNS/TLS observation only. This does not recreate the host environment or address selection at the failed requests and does not prove sustained connectivity."
|
||||
},
|
||||
"agSealedManifestUnchanged": true
|
||||
}
|
||||
@@ -1308,3 +1308,5 @@ AE两工程均真实自主交付通过,但Python内置检查选中了Hermes的
|
||||
原版Harness开放模式由collaborationInput将四项budget规范化为0;worker configFor跳过有限额度计算,engine credit以Infinity处理,capacityMode=passthrough不投影裁剪也不发输出cap。本机原生off/on各四请求和12000068合成tokens后继续完成父检查/Git,入口/apply、两个原生模块与进程树退出通过。夹具证明执行契约,不能代替真实自主交付/Purge收益。
|
||||
|
||||
AF正式规划零请求即拒绝,原因是Node绝对路径进入controlled/runtime身份:源码与正式二进制SHA相同也不代表身份相同。新AG必须以正式检测API为权威,隔离宿主用正式Node路径的四组夹具已通过。AF封存器把前置fixture结果对象误作隔离运行布尔值,原manifest已完整而公开索引缺失;离线逐SHA后仅补索引,新版仅fixture===true跳过公开索引。
|
||||
|
||||
AG四组中的前三组均connect ETIMEDOUT且无响应,第四关闭组8请求完整交付;总109888报告tokens+3unknown/44928本地估算不能相加当实际计费。两开启组原版入口/apply/两个原生模块已加载,工程未推进由网络失败终结,四例不等成功不算节省率。事后同正式Node二进制的新进程DNS和主机名/两IPv4的TLS均通过,不能复原失败时地址选择或证明长期稳定。原始失败、无型号可核对与unknown都应保留。
|
||||
|
||||
@@ -1946,3 +1946,5 @@ AE已封存225份52216677字节:React/Python均10请求、合计367365tokens/0
|
||||
AF准备完成:原生开放off/on四请求12000068合成tokens仍继续完成父交付;完整af-driver-c6ca1f0f四组16本机HTTP、原九测试/build/父Git/owner/index和原版入口/apply/模块/进程树全部通过,原状态审计及477份证据封存通过。旧函数名、Git断言、route-plan取证、null元数据和省略身份策略恢复失败全部保留。下一本地提交并唯一冻结AF真实ABBA;本机固定响应不算真实自主交付/Purge收益,不改产品和默认额度。
|
||||
|
||||
AF正式协作规划因源码Node/正式Node路径身份不同而拒绝,0真实请求/0worker;原50837行/118导出/全部工件和旧尝试保持,新增1自有父项目/会话/事件,池与配置恢复。71份30311984字节封存逐SHA通过,原manifest不改、公开索引离线补齐。AG另冻不续跑AF:首夹具TDZ失败保留,完整c5ff7602四组16自有HTTP、原九测试/build/父Git/原生与原状态审计通过,477份19575831字节封存,覆盖正式Node路径不同但二进制SHA相同。三系统接口只读反馈已记录,未扩大运行授权。下一本地提交后唯一冻结AG,真实收益仍unknown。
|
||||
|
||||
AG驱动/计划a6830bbf提交后唯一冻结并一次执行:前三组connect ETIMEDOUT无响应,第四关闭组8请求/109888tokens自主三源码与原9测试/build,自动父Git c792069c/finalize和owner/index通过。总11请求/109888报告tokens+3unknown/44928本地估算;两开启组原版入口/apply/两个原生模块已加载、四例进程树退出通过,四例不等成功,Purge收益仍unknown。原50841行/119导出/工件/旧尝试/凭据保持,新增8自有父子项目/会话,DeepSeek17→19/config118→120、池及其余配置恢复;final-36ea1aba审计与430份47723074字节封存通过。事后0HTTP/模型请求的DNS/三TLS观察通过,只证明该时点,原封存未改。正式1.25.8/PID71628未升级或重启;准备回顾/最终本地提交,后续Python解释器与终端Git报告另切修复。
|
||||
|
||||
@@ -0,0 +1,47 @@
|
||||
// Independent observation after AG is sealed. TLS only: no HTTP, model request or key access.
|
||||
import fs from 'node:fs';
|
||||
import path from 'node:path';
|
||||
import tls from 'node:tls';
|
||||
import { lookup, Resolver } from 'node:dns/promises';
|
||||
import { createHash, randomUUID } from 'node:crypto';
|
||||
import { fileURLToPath } from 'node:url';
|
||||
import assert from 'node:assert/strict';
|
||||
|
||||
const sealed = JSON.parse(fs.readFileSync('.qa/live-purge-ag/summary.json'));
|
||||
assert.equal(sealed.originalStateAuditResult.success, true);
|
||||
assert.ok(fs.existsSync('.qa/live-purge-ag/FINAL-MANIFEST.json'));
|
||||
const frozen = JSON.parse(fs.readFileSync('.qa/live-purge-ag/current.json'));
|
||||
const endpoint = new URL(frozen.connection.baseUrl);
|
||||
assert.equal(endpoint.protocol, 'https:'); assert.equal(endpoint.username, ''); assert.equal(endpoint.password, '');
|
||||
const hostname = endpoint.hostname, port = Number(endpoint.port || 443);
|
||||
const output = path.resolve('.qa/ag-network-' + randomUUID()); fs.mkdirSync(output);
|
||||
const source = fs.readFileSync(fileURLToPath(import.meta.url));
|
||||
fs.writeFileSync(path.join(output, 'observer.mjs'), source, { flag: 'wx' });
|
||||
const startedAt = new Date().toISOString(), resolver = new Resolver({ timeout: 5000, tries: 1 });
|
||||
const errorView = error => ({ name: error.name, code: error.code ?? null, syscall: error.syscall ?? null, message: error.message });
|
||||
const dnsResults = await Promise.allSettled([lookup(hostname, { all: true }), resolver.resolve4(hostname, { ttl: true }), resolver.resolve6(hostname, { ttl: true })]);
|
||||
const dns = Object.fromEntries(dnsResults.map((row, index) => [['system', 'ipv4', 'ipv6'][index],
|
||||
row.status === 'fulfilled' ? { success: true, values: row.value } : { success: false, error: errorView(row.reason) }]));
|
||||
const addresses = [...new Set([...(dns.system.values ?? []).map(row => row.address),
|
||||
...(dns.ipv4.values ?? []).map(row => row.address), ...(dns.ipv6.values ?? []).map(row => row.address)])];
|
||||
const probes = await Promise.all([{ label: 'hostname', host: hostname }, ...addresses.map(host => ({ label: 'resolved-address', host }))].map(target => new Promise(resolve => {
|
||||
const started = Date.now(); let settled = false, socket;
|
||||
const finish = value => { if (settled) return; settled = true; clearTimeout(timer); socket?.destroy(); resolve({ ...target, elapsedMs: Date.now() - started, ...value }); };
|
||||
const timer = setTimeout(() => finish({ success: false, error: { code: 'OBSERVATION_ENDED', message: 'This TLS observation ended after 8000 ms; no product setting was changed.' } }), 8000);
|
||||
try {
|
||||
socket = tls.connect({ host: target.host, servername: hostname, port, rejectUnauthorized: true, autoSelectFamily: true }, () => {
|
||||
const certificate = socket.getPeerCertificate();
|
||||
finish({ success: socket.authorized, remoteAddress: socket.remoteAddress, remoteFamily: socket.remoteFamily,
|
||||
protocol: socket.getProtocol(), certificateFingerprint256: certificate.fingerprint256 ?? null });
|
||||
});
|
||||
socket.once('error', error => finish({ success: false, error: errorView(error) }));
|
||||
} catch (error) { finish({ success: false, error: errorView(error) }); }
|
||||
})));
|
||||
const result = { schemaVersion: 1, startedAt, finishedAt: new Date().toISOString(), hostname, port,
|
||||
executable: process.execPath, executableSha256: createHash('sha256').update(fs.readFileSync(process.execPath)).digest('hex'),
|
||||
observerSha256: createHash('sha256').update(source).digest('hex'), dns, probes,
|
||||
proxyEnvironmentPresent: { http: !!process.env.HTTP_PROXY, https: !!process.env.HTTPS_PROXY, noProxy: !!process.env.NO_PROXY },
|
||||
httpRequests: 0, modelRequests: 0, keysReadOrResaved: false, productConfigurationChanged: false,
|
||||
scope: 'Fresh process DNS/TLS observation only. This does not recreate the host environment or address selection at the failed requests and does not prove sustained connectivity.' };
|
||||
fs.writeFileSync(path.join(output, 'result.json'), JSON.stringify(result, null, 2) + '\n', { flag: 'wx' });
|
||||
console.log(JSON.stringify({ output, ...result }));
|
||||
+3
-2
@@ -1,6 +1,6 @@
|
||||
# Agent Workbench:核心代理能力研究与产品计划
|
||||
|
||||
当前阶段AG:AD Node与AE React/Python真实自主交付及原证据封存已完成;AF在正式规划时因源码/宿主Node路径身份不符而0请求终止,71份原件及原状态审计封存。AG完整隔离驱动c5ff7602四组16自有HTTP/原九测试/父Git与恢复审计通过,正式Node路径差异已覆盖。下一提交驱动和计划后唯一冻结AG、一次执行原版Purge ABBA并审计封存;收益仍unknown。不改正式产品、不重放AF/旧样本,不代交并行品牌/UI改动。见docs/QA-LIVE-PURGE-AG.md。
|
||||
当前阶段AG已完成执行/审计/封存:11真实请求、109888报告tokens+3unknown/44928本地估算。前三组connect超时,ba-off8请求自主源码/原9测试/build、自动父Git c792069c/finalize与owner/index通过;开启组原版加载/apply/两个原生模块与四例进程退出核对,四例不等成功,Purge收益仍unknown。原50841行/119导出/工件/尝试/凭据保持、池与配置恢复,430份47723074字节封存。AF0请求原失败/71份封存不改;AG隔离夹具及正式API身份解决Node路径摘要差异。下一独立产品切片修Python检查解释器选择和终端Git/结束报告登记;不重放旧样本,不代交或构建并行品牌/UI。见docs/QA-LIVE-PURGE-AG.md。
|
||||
|
||||
## 当前接续:C 工具可见性、前缀证据与新真实验证(进行中)
|
||||
|
||||
@@ -1020,4 +1020,5 @@ F收尾:57d075e修正独立编译QA路径,6d09ac8封存差异;202610302320
|
||||
4. [complete] AD驱动/真实兼容与独立Node已封存:21请求199250报告tokens+2unknown,Node11请求9/9/build/Git88bb25cb及owner/index通过。
|
||||
5. [complete] AE React/Python均自主交付,原九/十测试、类型/构建/页面、限定Git与owner/index通过;20请求367365tokens/0unknown,225份证据和原数据审计封存。
|
||||
6. [complete] 开放worker/engine的0不限契约与完整四组本机驱动已通过;AF正式规划失败0请求,原失败、71份封存与审计保持。
|
||||
7. [in_progress] 新AG正式宿主API身份完整夹具通过;提交并唯一冻结后执行原版Purge AB/BA,真实收益仍unknown。
|
||||
7. [complete] AG原版Purge AB/BA一次执行/审计/430份封存通过:1工程成功、3网络unknown,收益未验证;事后0HTTP的DNS/三TLS通过不改写原失败。
|
||||
8. [pending] 独立修Python检查解释器选择与终端Git/结束报告衔接;三系统先只读工程证据对接,入站版本化/幂等服务尚未实现。
|
||||
|
||||
Reference in New Issue
Block a user