docs: seal Messages release and rejected Claude identity preflight

This commit is contained in:
Agent Workbench Dev
2026-11-03 20:14:37 +08:00
parent d62d03f93f
commit 860a191b8b
759 changed files with 620129 additions and 6 deletions
+12
View File
@@ -182,3 +182,15 @@ docs/apikey-route-l-evidence/**/*.body binary
docs/apikey-route-l-evidence/observations/public-main.js binary
docs/apikey-route-l-offline.txt binary
docs/apikey-route-l-offline-final.txt binary
docs/messages-m-evidence/** -text whitespace=blank-at-eol,blank-at-eof,space-before-tab,cr-at-eol
docs/messages-m-evidence/reference/** binary
docs/messages-m-evidence/source-bytes/** binary
docs/messages-m-evidence/**/*.log binary
docs/messages-m-evidence/**/*.txt binary
docs/messages-m-evidence/**/*.sse binary
docs/messages-m-evidence/**/*.tap binary
docs/messages-m-evidence/**/*.body binary
docs/messages-m-offline.txt binary
docs/messages-m-git-index.txt binary
docs/anthropic-messages-release-identity.json -text
+7 -1
View File
@@ -1,5 +1,11 @@
# Agent Workbench项目执行规范
M已本地交付:1.24.4/source d62d03f9/候选20261103112054864,原生Messages、两种显式鉴权、实际URL、工具/PNG历史、JSON/SSE/身份与缓存用量已接入。25专项/144相关及61+48原协议、18升级相关、549源码ZIP输入重建22/22、三归档/便携、源码/便携/安装载荷各七组UI、4类历史UI、12实际安装/16+9安装UI通过。正式保留PID36552/Provider,原55项目71会话及全原行/导出/3570原工件/公开配置精确保持,网页/原生及再次核对通过。745份/54940236字节证据离线通过,见docs/QA-ANTHROPIC-MESSAGES-M.md及docs/anthropic-messages-release-identity.json;首兼容清单、封存取错目录和UI操作拒绝保持,不重建已安装包。
M真实单次预检已结束:同apikey原凭据修订2,Messages/Bearer实际https://api.apikey.fan/v1/messages请求claude-opus-5-5,但原message_start报告gpt-6.1-sol;1请求/221报告tokens/0未知/0工具/无重试切池,strict暂停。原201尝试、71原导出/3570原工件及所有原行按精确声明归一化保持,新增1自有项目/1会话/8记录/22事件与outbox/1尝试/4工件,当前56项目72会话。目标连接保留更正协议/严格身份,修订8→9;全局配置/默认高性能池保持,9229关闭,无Key提取或重存。Claude线路身份仍未通过,不能添加无依据别名、不重跑J/E/H/M,不把协议验收或返回型号当权重认证/工程/Purge收益。下一先核对供应商路由或新选连接,再独立单次预检,通过后按J计划做新Node工程与单列Purge。
下方M首候选和开发阶段记录为历史时点,不代表当前正式状态。
M首候选65ddee16/20261103105910315未部署:三归档、549源码输入/22编译独立重建、便携和七组UI通过,实际隔离安装第六组被热升级入口漏列1.24.4拦下;补齐后另制新不可变候选,不放宽故障注入期望。原正式1.24.3/PID36552/Provider与凭据保持,真实预检尚未发出。保留首候选和首失败,后续以最终M发行身份为准。
M原生Messages开发验收中:1.24.4源码已接入Anthropic请求/工具与PNG历史、JSON/SSE/累计缓存用量、起始型号、严格暂停、两种显式鉴权与实际URL。首23专项、61相关、48旧协议/身份及编译宿主七组两视口新UI/四类历史UI通过;首307诊断漏URL已修正,原失败保留。当前正式仍原1.24.3/PID36552/Provider;新发行独立源码/便携/安装、保留状态正式升级与单次真实Messages身份预检尚未完成。不得把夹具验证当Claude线路已修复或真实工程/Purge通过;原J/K/L证据保持。
@@ -44,7 +50,7 @@ D固定可信来源生产策略已完成:实现9106655/最终验收d05a741,6
## 范围与目标
当前接续优先级:G已完成用户要求的已确认问题修复、正式部署和原生窗口回看。后续真实模型自主交付/Purge收益须新开有界样本、分别记录,不重跑旧失败或将未交付对照宣称有收益;docs/PLAN-1.25.md及新iPhone/其它供应商方案保留。发布前多轮完整目录核验耗时已记录为后续性能审查,不在运行中的安装流程削弱核验或改写载荷。
当前接续优先级:M原生Messages已完成本地交付,单次真实身份预检仍不符;先核对供应商Claude路由或用户明确新选连接,再按J计划另开单次预检。通过后才进入新Node工程和原版Purge独立实测,不重跑旧失败或将未交付对照宣称收益。docs/PLAN-1.25.md及新iPhone/其它供应商方案保留。发布前多轮完整目录核验耗时已记录为后续性能审查,不在运行中的安装流程削弱核验或改写载荷。
本仓库专注AI编程与能力研究工作台。其他项目不纳入本项目计划或代码。用户确定的研究对象为Codex、Claude Code、Cursor、DeepSeek Harness、WorkBuddy及DSH-purge;目标是将有证据的机制独立开发成自有Harness内置能力。
+2 -1
View File
@@ -1,11 +1,12 @@
# Agent Workbench 更新记录
## 1.24.4 · 原生 Anthropic Messages(开发验收中)
## 1.24.4 · 原生 Anthropic Messages(已本地交付)
- 模型中心增加 Messages 协议和 x-api-key/Bearer 鉴权;预览最终请求地址,支持服务根或 /v1 根地址,保存切换不自动调用模型。
- 接入原生系统/工具/图片历史、JSON/SSE、停止与完整性检查,以及起始事件中的真实返回型号;不完整或身份不符时保留原响应并暂停。
- 保留原生用量和缓存分项,累计流用量只结算一次;账本和错误显示请求/响应地址,模型请求不自动跟随重定向。
- 本机协议、界面与发行验收分别记录;真实连接身份及工程/Purge效果单独判定,未完成项目不宣称已通过。
- 独立源码重建、三归档、便携和实际安装、正式网页/原生回看及原数据保留通过。原apikey一次Messages预检仍返回GPT标识,已严格暂停并保留221报告tokens与原响应;Claude线路身份、真实工程与Purge收益尚未通过,见docs/QA-ANTHROPIC-MESSAGES-M.md。
## 1.24.3 · 统一模型选择入口(已本地交付)
+2
View File
@@ -1,5 +1,7 @@
# J:I交付后的新真实样本计划(单次身份预检未通过)
M接续状态:1.24.4已本地交付原生Messages,原apikey更正为Messages/Bearer后实际/v1/messages单次预检仍报告gpt-6.1-sol,请求为claude-opus-5-5;1请求/221报告tokens/0工具,strict暂停。见[原生协议验收与新结果](QA-ANTHROPIC-MESSAGES-M.md)。先核对供应商路由或新选连接,再新开下列步骤1;不能自动重试M/J或以别名强行通过。下列工程与Purge仍待真实身份预检通过。
用户选择 apikey Claude 后已执行一次独立真实预检:请求 claude-opus-5-5,上游完整 SSE 标记 gpt-6.1-sol;1 请求/4334 报告 tokens/0 工具,严格暂停。原参数/号池与凭据保持,原状态核对完成;见[实测结果](LIVE-MODEL-IDENTITY-J-RESULTS.md)。下列工程与 Purge 尚未执行,需更正或新选连接后另开预检;不自动重试本样本。
1. 新开独立身份预检:冻结已配置连接和版本,严格核对请求/返回身份及来源;一请求、最多256输出、累计16000tokens、无重试或切池。输入估算超过原限即零发送拒绝。身份缺失、不一致、混合或中断立即停止,映射须有实际依据。凭据仅由原Provider使用。
+80
View File
@@ -0,0 +1,80 @@
# 1.24.4:原生 Messages 已本地交付,Claude 线路身份仍不一致
工作台已在 `http://127.0.0.1:4318` 升级为 1.24.4。正式宿主仍为 PID36552,原 Provider 与凭据保持;原55项目、71会话、所有原表行、完整导出、3570份原工件文件及模型中心在升级和窗口回看后精确一致。
随后同一 `apikey` 连接仅发送了一次新的原生 Anthropic Messages 请求。实际请求和响应地址均为 `https://api.apikey.fan/v1/messages`,请求型号 `claude-opus-5-5`,上游原始 `message_start.message.model` 仍为 `gpt-6.1-sol`。严格身份策略已暂停该新任务,0工具执行、0重试、0切池;上游报告192输入、29输出、合计221 tokens,无未知用量。**产品协议交付通过,Claude 线路身份预检未通过。**
## 产品修复
L调查已确认旧版只有 Chat/Responses,供应商公开 Claude 示例使用 `/v1/messages`;不能仅改 URL 后继续发送 Chat 格式。1.24.4 增加独立 `anthropic-messages` 协议,支持服务根或 `/v1` 根地址、显式 `x-api-key`/Bearer 鉴权,预览最终请求地址。
适配原生 system、文本/PNG、tool_use/tool_result 历史、保留的 thinking 签名、JSON/SSE 完整终结及起始事件型号。累计流用量只结算一次,输入包含缓存创建/读取分项;支持的原生参数按协议发送。混合协议池保留每条线路身份,错误及请求账本保留实际请求/响应 URL、原响应、原请求ID和用量。模型请求不自动跟随 HTTP 重定向。
型号不符时保留完整响应并暂停,不执行其中工具,也不把 GPT 加入 Claude 的允许别名。协议和界面夹具通过不能证明供应商的真实模型、工程交付或 Purge 收益。
## 不可变发行与验收
最终来源为 `d62d03f93f0c1a94c3475bcc14385c4a3d9cde7a`,候选 `Agent-Workbench-1.24.4-win-x64-20261103112054864`。安装根 `C:/Users/wangy/AppData/Local/Programs/AgentWorkbench`。后续报告/离线验证器提交不重新制作已安装载荷;完整身份见 [发行记录](anthropic-messages-release-identity.json)。
| 范围 | 结果与证据 |
| --- | --- |
| Messages专项与关联回归 | 最终25专项、144相关、此前61相关/48旧协议身份通过;[专项日志](messages-m-evidence/development/specialist-final.log)、[相关日志](messages-m-evidence/development/final-related.log) |
| 升级兼容修复 | 构建及18相关升级/回退/安装检查通过;[日志](messages-m-evidence/development/upgrade-fix-tests.log) |
| 新界面流程 | 最终源码、便携、正式安装载荷的隔离宿主各7组,含390px、工具往返、缓存用量、严格不符及429原错误,页面错误0;[源码](messages-m-evidence/cases/sourceUI/result.json)、[便携](messages-m-evidence/cases/portableUI/result.json)、[安装载荷](messages-m-evidence/cases/installedUI/result.json) |
| 相关历史界面 | 模型中心、协议、身份、模型选择4类通过;[收据](messages-m-evidence/cases/regression/result.json) |
| 独立构建和发行 | 冻结544产品输入/22编译文件;源码ZIP549输入含5个示例生成文件,离线重建22/22逐字节一致;三归档CRC/目录/清单/原字节及实际便携通过;[冻结](messages-m-evidence/release/product-freeze.json)、[重建](messages-m-evidence/release/rebuild.json)、[归档](messages-m-evidence/release/archives.json) |
| Windows实际安装 | 12组生命周期通过,含活动锁定、故障回退、原进程/凭据保留、真实重启恢复、卸载留数据,安装副本16+9组流程通过;[收据](messages-m-evidence/release/installation.json) |
| 正式升级和回看 | 原PID/Provider、55项目71会话及所有原行/导出/工件/公开配置精确保持;正式网页和原生窗口实际回看后再次核对通过,9229关闭;[最终核对](messages-m-evidence/verification/formal/final.json)、[网页](messages-m-evidence/verification/browser/result.json)、[原生窗口](messages-m-evidence/verification/native/result.json) |
| 单次预检驱动 | 自有localhost匹配/不符/429三场景均1请求0工具,调试端口关闭;与最终220产品src输入逐字节绑定;[收据](messages-m-evidence/cases/probe-driver/result.json)、[绑定](messages-m-evidence/cases/probe-driver/source-binding.json) |
这是相关范围验收,没有把全部历史后端整轮或新iPhone实机验证记成通过。表中独立宿主模型请求均为自有localhost夹具;只有下面一条正式预检调用真实配置服务。
## 真实预检原件与边界
| 项目 | 原记录 |
| --- | --- |
| 连接 | `apikey` / `09b0293b-faad-45c3-832d-7c8b615813aa`,同一原Provider凭据修订2 |
| 协议/鉴权 | `anthropic-messages` / Bearer / anthropic-version `2023-06-01` |
| 请求/响应地址 | `https://api.apikey.fan/v1/messages` |
| 请求型号 | `claude-opus-5-5` |
| 原始上游型号 | `gpt-6.1-sol`,来自 `message_start.message.model` |
| 限制 | 1请求,max_tokens256,累计预算16000,实际tools字段缺省且工具定义/执行均0;不重试、不切池 |
| 结果 | `mismatch` / `paused` / `model_identity`,完整流终结,192+29=221报告tokens |
| 新会话 | `f7d1ec3d-aa41-4f4e-ab6d-f4df487b4d76` |
| 新Run | `b175e0e0-1472-42af-99dd-bda5d1edcc0b` |
| 请求SHA256 | `5420746a315e7226d1636d6229bdb4545b2089c3f3e36c2019bfce32e250ef40` |
| 响应SHA256 | `403a817ff9c9b4d4f7bd0094eac561cd358a541398a4875c9167a08ece1482c4`,3214字节 |
[原契约](messages-m-evidence/live/contract.json)、[实际请求](messages-m-evidence/live/request.json)、[原始SSE](messages-m-evidence/live/response.sse)、[结果](messages-m-evidence/live/result.json)、[完整新会话导出](messages-m-evidence/live/export.json)均已封存。正式账本实际回看显示相同请求/响应地址、Claude请求和GPT上游型号;[截图](messages-m-evidence/verification/browser/live-ledger.jpg)。
这次证据说明:在正确原生路径和原生请求体下,原连接仍报告GPT;只补协议不能解决现有线路的型号差异。本次没有 legacy 请求、池内切换或客户端默认GPT回填。服务端映射、Key分组和底层权重仍未知,不能仅凭返回字符串断言实际运行的权重,也不能给它添加无依据别名绕过核对。旧J/E/H原件和失败结论保持。
本次预检由一次性QA驱动调用原进程中的 Provider、ModelHub 和 RequestPlanner,并记录所属Task/Run、请求工件和原用量;真正移除了工具定义。它是连接身份测试,未运行完整自主工程循环,未调用工具、检查、Git交付或原版Purge。
## 正式状态保留
预检审计核对71个原完整导出、3570个原工件文件以及全部原表行,原201条尝试精确保留。只新增1个自有项目、1个自有会话、8记录、22事件和对应22条outbox、1条请求尝试、4个工件文件;原55项目位置严格加1,事件序列按所属追加增长。当前为56项目、72会话、6094记录、12953事件。
只将目标连接从Chat改为Messages/Bearer、原生默认参数和严格身份策略,连接修订8→9;凭据修订仍2,没有读取、复制、解密或重存Key。全局配置、默认「高性能」池及其他连接保持;原磁盘凭据块在本轮前已是source_changed,本轮没有把它冒充可重启恢复。[精确审计](messages-m-evidence/live/state-audit.json)和[端口关闭收据](messages-m-evidence/live/inspector-final.json)可复核。
原连接保留更正后的协议与严格核对,不恢复错误的Chat配置,也不改变默认池;当前该连接显示「需要处理」。未自动选择其他模型,未重跑J/E/H或任何用户旧任务。
## 失败回顾与下一顺序
首候选 `65ddee16/20261103105910315` 的归档、独立重建、便携及界面已通过,但实际安装第六组在指针故障注入之前被兼容清单拦下:`upgrade-running.mjs` 漏列1.24.4。该候选从未正式部署。补齐清单、通过18相关检查后制作最终新候选,最终12组安装不放宽故障期望;[首候选说明](messages-m-evidence/release/first-candidate-review.json)、[原失败](messages-m-evidence/release/first/installation.json)保留。
开发首307诊断漏URL失败已修复。预检夹具首两轮误将重复进度事件计为多个请求,改为按wireAttemptId去重后通过,原日志/驱动保留。正式原生启动器窗口发现延迟、漏scrollX的拒绝和滚动后UIA坐标越界均按新观察恢复,没有重启宿主或保存连接;网页浏览器自动填写的未保存表单已取消,未取出其凭据值。
首封存脚本错误地从源码目录读取编译输出而拒绝,首脚本/日志和部分目录保持;修正为从已验不可变便携目录读取22份输出,仍逐项要求与冻结SHA一致,未重建或改写发行。实际安装737175ms,多轮完整性核验耗时继续列为后续性能研究,不在运行中的升级弱化校验。
下一优先核对供应商这条凭据所对应的Claude模型路由或使用用户明确选定的可靠连接。可在本机对照本报告的原始请求/响应,请勿通过聊天提供Key。尚未替用户联系供应商,也不继续消耗这条线路反复试探。
线路明确更正后,另建一次同等有界身份预检;通过后再按 [真实任务计划](LIVE-RETEST-J-PLAN.md) 做新的Node工程交付,核对文件、实际检查、审查和Git。原版Purge按固定输入、新工程、AB/BA及同一真实连接单独实测,不将协议适配、提示或失败提前停止当收益。1.25其他扩展继续保留为计划。
## 证据
745份原件共54940236字节,manifest SHA256为 `c35590ff7879807e9302dc25ab24bf246b97396c9c78167e981c222db722e4c7`。排除Key/加密凭据块、宿主授权秘密、私有数据库和完整原用户快照;失败原件保持。
证据及清单746份/55074620字节已在实际Git索引逐项核对长度与SHA256一致。首次cached diff检查指出原CRLF收据和供应商HTML尾空格;保留原件,只调整本轮证据的字节/whitespace属性,不修剪第三方或捕获数据。
`node scripts/verify-messages-m.mjs` 离线核对原字节、发行和重建绑定、界面/安装/正式保留、单次无工具请求及原SSE型号,不访问网络或模型。见 [离线结果](messages-m-offline.txt)、[证据清单](messages-m-evidence/manifest.json)、[Git索引原字节核对](messages-m-git-index.txt)。验证器通过表示证据一致和受限流程正确,**不表示Claude身份预检通过**。
+2
View File
@@ -1,5 +1,7 @@
# Agent Workbench 验收记录
1.24.4已本地交付原生Anthropic Messages与实际URL诊断。最终d62d03f9/候选20261103112054864,25专项/144相关、源码ZIP549输入独立重建22/22、三归档/便携、三载荷各七组UI、12实际安装及正式网页/原生回看通过,原PID36552/Provider与55项目71会话所有原数据精确保持。随后唯一真实Messages预检仍返回GPT标识,1请求/221报告tokens/0工具,严格暂停;当前56项目72会话追加均归属该预检,原数据和默认池保持。745份证据离线核对通过不表示Claude线路身份通过,见[完整报告](QA-ANTHROPIC-MESSAGES-M.md)及[发行身份](anthropic-messages-release-identity.json)。真实工程/Purge与1.25扩展保持待验/计划,下方按历史版本理解。
1.22.0已完成所属Vite预览、独立桌面浏览区、有界DOM/原截图与审批动作、独立登录资料/撤销/下载、手机原观察SHA核对的本地交付。独立源码683/683(C完整原轮通过,D严格绑定后沿用)、33类250组原B串行界面(产品输入相同,严格绑定后沿用)、源码/便携/实际安装各17组浏览流程、同载荷安装7组、正式30页及Edge/Electron两视口通过。原PID36552/Provider保持,17项目/32全部会话/3239记录/6229事件及1780工件文件原字节保持,9229关闭。本轮验收发起真实模型请求0。
见[使用说明](EMBEDDED-BROWSER.md)、[1.22验收](QA-1.22.md)、[回顾](RETROSPECTIVE-1.22.md)与[发行身份](embedded-browser-release-identity.json)。下一[1.23发行效率与更新进度](PLAN-1.23.md)仅方案;真实模型自主工程/Purge收益和新iPhone流程另验。
@@ -0,0 +1,242 @@
{
"schemaVersion": 1,
"version": "1.24.4",
"sourceCommit": "d62d03f93f0c1a94c3475bcc14385c4a3d9cde7a",
"candidate": "Agent-Workbench-1.24.4-win-x64-20261103112054864",
"portableRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103112054864",
"installedRuntimeRoot": "C:\\Users\\wangy\\AppData\\Local\\Programs\\AgentWorkbench\\versions\\Agent-Workbench-1.24.4-win-x64-20261103112054864",
"installer": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103112054864-Setup.exe",
"bytes": 213126171,
"sha256": "4012a7914c4e0f44f93c56d6603c9c2000b3090971fee2a688855de3c8c94960",
"compiler": {
"version": "6.3.1",
"sha256": "3b22d36fc7e7c421d8caf780008ecdb42b14b3f0371b9bafe3cad64e6648d9cf"
},
"source": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\installer-build-f7b3301c-b38c-4143-aaee-d91718c3c75a\\installer.iss",
"version": "1.24.4",
"candidateDirectory": "Agent-Workbench-1.24.4-win-x64-20261103112054864",
"qaFixture": null
},
"archives": {
"portable": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103112054864.zip",
"bytes": 242709419,
"sha256": "31983cb4258ebcf517ee66f5da2b4e0464cc0e04620341a2491b38b5d56ce329"
},
"source": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-source-20261103112054864.zip",
"bytes": 894781411,
"sha256": "20c6c851480abaec78de0860885f41370e27773863ae96d0dac97f0e437d51d0"
},
"evidence": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-evidence-20261103112054864.zip",
"bytes": 653648820,
"sha256": "3b3ea76e98739b657fda91bdb1d95e60f451346f9d6453125a53aa6dc7ae4ef2"
}
},
"documentation": {
"sourceCommit": "d62d03f93f0c1a94c3475bcc14385c4a3d9cde7a",
"originalFiles": 18877,
"originalBytes": 2026294838,
"shippedFiles": 141,
"shippedBytes": 1091320,
"evidenceManifestSha256": "f3dfb490d57fd60e7891ed347a31bf4afeab5649798138a370c11bd4c32fd937"
},
"formalBeforeProbe": {
"url": "http://127.0.0.1:4318",
"pid": 36552,
"hasKey": true,
"projects": 55,
"sessions": 71,
"retention": {
"originalRows": "exact",
"originalExports": "exact",
"originalArtifactBytes": "exact",
"originalModelCenter": "exact"
},
"desktopWindows": 1,
"inspectorClosed": true,
"manifestSha256": "a7a0a64bda8babe16bec85b9f1b5e195291779b79db8cff478e3a9fe123e8c2f",
"runtimeIdentity": "af61c35c5465880fd73fb76a597972286fe49f8fd6a3d2cd8ac706845d05c00d"
},
"formalAfterProbe": {
"version": "1.24.4",
"pid": 36552,
"projects": 56,
"sessions": 72,
"tables": {
"events": {
"count": 12953,
"sha256": "fc8d3b5f53e8b62ca29a1278962e7cc0914ab46a19f044fa92a53745b966d007"
},
"metadata": {
"count": 6,
"sha256": "35e85bc531027a00c7364e24135d8c1d4c6053f6d962ac226a00ee563bf92d45"
},
"outbox": {
"count": 12953,
"sha256": "0fdabf07b4e002792e484e3d7dd70234fc2584a4455f40e26e2e0cc6205bd9e1"
},
"projects": {
"count": 56,
"sha256": "7edec7632c6fae3d51fa54d09bf16976492177b75160f851dec10b76b14c9411"
},
"records": {
"count": 6094,
"sha256": "6d938fe9a7956f8aa6545b133067048ab7f9a52c6ec7291b7af2b302dbbd0f62"
},
"sessions": {
"count": 72,
"sha256": "2e9ce85c08b42213432d8337b322538165b4f15e0a34f5e7c42aa2f9b5bb999a"
},
"sqlite_sequence": {
"count": 1,
"sha256": "a09d302adfdb6c3361b92d9249cbbec167ec0125c5fd064205707d7f17bb6739"
}
},
"originalExportsExact": 71,
"originalArtifactFilesExact": 3570,
"originalRowsAndDeclaredChanges": [
{
"table": "events",
"original": 12931,
"current": 12953,
"exact": 12931,
"declared": 0
},
{
"table": "metadata",
"original": 6,
"current": 6,
"exact": 4,
"declared": 2
},
{
"table": "outbox",
"original": 12931,
"current": 12953,
"exact": 12931,
"declared": 0
},
{
"table": "projects",
"original": 55,
"current": 56,
"exact": 0,
"declared": 55
},
{
"table": "records",
"original": 6086,
"current": 6094,
"exact": 6086,
"declared": 0
},
{
"table": "sessions",
"original": 71,
"current": 72,
"exact": 71,
"declared": 0
},
{
"table": "sqlite_sequence",
"original": 1,
"current": 1,
"exact": 0,
"declared": 1
}
],
"globalConfigAndSelectionExact": true,
"noCredentialExtractionOrResave": true,
"credentialRevision": 2,
"connectionRevision": 9
},
"preflight": {
"completedAt": "2026-11-03T12:03:16.397Z",
"version": "1.24.4",
"pid": 36552,
"connectionId": "09b0293b-faad-45c3-832d-7c8b615813aa",
"sessionId": "f7d1ec3d-aa41-4f4e-ab6d-f4df487b4d76",
"taskId": "15d22836-27a0-422c-a0a3-d0ed1752d1f4",
"runId": "b175e0e0-1472-42af-99dd-bda5d1edcc0b",
"protocol": "anthropic-messages",
"messagesAuth": "bearer",
"requestUrl": "https://api.apikey.fan/v1/messages",
"responseUrl": "https://api.apikey.fan/v1/messages",
"requestedModel": "claude-opus-5-5",
"reportedModels": [
"gpt-6.1-sol"
],
"identityStatus": "mismatch",
"identityPassed": false,
"sessionStatus": "paused",
"stopReason": "model_identity",
"requests": 1,
"toolDefinitions": 0,
"actualTools": 0,
"usage": {
"prompt_tokens": 192,
"completion_tokens": 29,
"total_tokens": 221,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 192,
"output_tokens": 29,
"cache_creation_input_tokens": 0,
"cache_read_input_tokens": 0
},
"protocol": "anthropic-messages"
},
"usageStatus": "reported",
"runUsage": {
"requests": 1,
"inputTokens": 192,
"outputTokens": 29,
"totalTokens": 221,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"captures": [
{
"kind": "request",
"file": "request.json",
"id": "8f508263-0025-4553-a194-1daff505fc22",
"bytes": 440,
"sha256": "5420746a315e7226d1636d6229bdb4545b2089c3f3e36c2019bfce32e250ef40"
},
{
"kind": "response",
"file": "response.sse",
"id": "eeb6c2ef-c9aa-4202-ab19-dbc19877d0bc",
"bytes": 3214,
"sha256": "403a817ff9c9b4d4f7bd0094eac561cd358a541398a4875c9167a08ece1482c4"
}
],
"noRetry": true,
"noPoolSwitch": true,
"modelWeightsAuthenticated": false,
"realEngineeringDeliveryTested": false,
"originalPurgeBenefitTested": false,
"inspectorClosed": true,
"defaultSelectionRetained": true,
"credentialRevisionRetained": true
},
"evidence": {
"manifest": "docs/messages-m-evidence/manifest.json",
"sha256": "c35590ff7879807e9302dc25ab24bf246b97396c9c78167e981c222db722e4c7",
"files": 745,
"bytes": 54940236,
"offline": "docs/messages-m-offline.txt",
"gitIndex": "docs/messages-m-git-index.txt",
"report": "docs/QA-ANTHROPIC-MESSAGES-M.md"
},
"originalJManifestSha256": "4ac0aaa83c4aee37a6f514d9a6021ec92830d6ad464c7f9fb39f331561d8754f",
"originalLManifestSha256": "bd900d169944161058e25fa4fa2675560e44e3f36e9b73ef8dab14078e111fad",
"realModelRequests": 1,
"scope": "Native Messages product release accepted. The single real Claude-request preflight was rejected because the upstream native stream reported GPT. No retry, tool dispatch, credential extraction/resave, underlying-weight authentication, engineering-delivery or Purge-benefit assertion. Final documentation does not replace the immutable deployed package."
}
Binary file not shown.

After

Width:  |  Height:  |  Size: 138 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 141 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 40 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 49 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 58 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 55 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 53 KiB

File diff suppressed because one or more lines are too long
Binary file not shown.

After

Width:  |  Height:  |  Size: 138 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 141 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 40 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 50 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 58 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 55 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 53 KiB

File diff suppressed because one or more lines are too long
@@ -0,0 +1,178 @@
{
"id": "41c28fcb-4300-4005-bc2c-4b7fce3b2bf3",
"requestId": "3c21957a-6ce7-4682-a24e-0212eace4a2a",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"connectionId": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"connectionRevision": 1,
"credentialRevision": 1,
"poolId": null,
"poolRevision": 1,
"model": "claude-owned-m",
"protocol": "anthropic-messages",
"streaming": true,
"endpoint": "http://127.0.0.1:49372",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"messagesAuth": "bearer",
"ordinal": 1,
"status": "failed",
"startedAt": "2026-11-03T11:07:01.846Z",
"requestBytes": 320,
"requirements": [
"text"
],
"rejections": [],
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"usage": null,
"usageStatus": "unknown",
"requestArtifactId": "03055e60-b6c0-47e9-ac4b-60a2436989c3",
"wireAttemptId": "a4805e41-b1b1-4aaf-9533-8b230f3878ae",
"upstreamResponse": {
"httpStatus": 429,
"requestId": "owned_m_429",
"headers": {
"content-type": "application/json",
"retry-after": "1",
"date": "Tue, 03 Nov 2026 11:07:02 GMT",
"request-id": "owned_m_429"
},
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"responseUrl": "http://127.0.0.1:49372/v1/messages",
"redirected": false,
"redirectMode": "manual"
},
"responseArtifactId": "61498b0d-0814-4aed-9ce0-c32433221d98",
"responseSha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3",
"responseRedacted": false,
"upstream": {
"schema": 1,
"scope": "http-body",
"httpStatus": 429,
"code": null,
"type": "rate_limit_error",
"message": "Owned preflight rate limit",
"param": null,
"reason": null,
"requestId": "owned_m_429",
"responseId": null,
"eventType": null,
"headers": {
"content-type": "application/json",
"retry-after": "1",
"date": "Tue, 03 Nov 2026 11:07:02 GMT",
"request-id": "owned_m_429"
},
"retryAfter": "1",
"retryAfterMs": 1000,
"rawPreview": "{\"type\":\"error\",\"error\":{\"type\":\"rate_limit_error\",\"message\":\"Owned preflight rate limit\"}}",
"previewTruncated": false,
"capture": {
"scope": "http-body",
"receivedBytes": 91,
"capturedBytes": 91,
"storedBytes": 91,
"retention": "artifact",
"complete": true,
"truncated": false,
"interrupted": false,
"capturedRange": [
0,
91
],
"redacted": false,
"sha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3"
},
"artifactId": "61498b0d-0814-4aed-9ce0-c32433221d98",
"artifactSha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3",
"artifactScope": "http-body",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40"
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [],
"status": "incomplete",
"complete": false,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights."
},
"diagnostic": {
"id": "42238d3b-0f8b-4256-b188-2aef8ec619ca",
"source": "model_connection",
"origin": "upstream",
"stage": "model_request",
"kind": "rate_limit",
"httpStatus": 429,
"endpoint": "http://127.0.0.1:49372",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempts": 1,
"occurredAt": "2026-11-03T11:07:02.336Z",
"retryable": true,
"title": "模型服务请求达到限额",
"upstream": {
"schema": 1,
"scope": "http-body",
"httpStatus": 429,
"code": null,
"type": "rate_limit_error",
"message": "Owned preflight rate limit",
"param": null,
"reason": null,
"requestId": "owned_m_429",
"responseId": null,
"eventType": null,
"headers": {
"content-type": "application/json",
"retry-after": "1",
"date": "Tue, 03 Nov 2026 11:07:02 GMT",
"request-id": "owned_m_429"
},
"retryAfter": "1",
"retryAfterMs": 1000,
"rawPreview": "{\"type\":\"error\",\"error\":{\"type\":\"rate_limit_error\",\"message\":\"Owned preflight rate limit\"}}",
"previewTruncated": false,
"capture": {
"scope": "http-body",
"receivedBytes": 91,
"capturedBytes": 91,
"storedBytes": 91,
"retention": "artifact",
"complete": true,
"truncated": false,
"interrupted": false,
"capturedRange": [
0,
91
],
"redacted": false,
"sha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3"
},
"artifactId": "61498b0d-0814-4aed-9ce0-c32433221d98",
"artifactSha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3",
"artifactScope": "http-body",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40"
},
"upstreamRequestId": "owned_m_429",
"code": null,
"type": "rate_limit_error",
"retryAfterMs": 1000,
"limitReason": "rate",
"wireAttemptId": "a4805e41-b1b1-4aaf-9533-8b230f3878ae",
"message": "模型服务请求达到限额(HTTP 429)。 Owned preflight rate limit",
"detail": "Owned preflight rate limit",
"suggestion": "依据上游错误码核对限制原因;有 Retry-After 时按原等待时间处理,原因未给出时保持未确定。",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"responseUrl": "http://127.0.0.1:49372/v1/messages",
"redirected": false,
"redirectMode": "manual",
"retryDecision": "attempt_limit",
"remainingWaitBudgetMs": 299634
},
"reason": "模型服务请求达到限额(HTTP 429)。 Owned preflight rate limit",
"httpStatus": 429,
"finishedAt": "2026-11-03T11:07:02.495Z"
}
@@ -0,0 +1,785 @@
{
"schemaVersion": 1,
"executionSchemaVersion": 1,
"exportedAt": "2026-11-03T11:07:02.989Z",
"project": {
"id": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\429\\project",
"name": "project",
"addedAt": "2026-11-03T11:07:01.366Z"
},
"session": {
"id": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"title": "Owned M probe 429",
"status": "failed",
"createdAt": "2026-11-03T11:07:01.397Z",
"updatedAt": "2026-11-03T11:07:02.719Z",
"messages": [
{
"id": "d85b47c8-790f-4b77-8362-52a3835060a3",
"role": "user",
"content": "Reply OK owned-m. Do not use tools.",
"createdAt": "2026-11-03T11:07:01.636Z"
},
{
"id": "fbbe79d4-fa56-4a78-9f84-c5d53d50be76",
"role": "system",
"content": "模型服务请求达到限额(HTTP 429)。 Owned preflight rate limit",
"createdAt": "2026-11-03T11:07:02.528Z"
}
],
"providerMessages": [
{
"role": "user",
"content": "Reply OK owned-m. Do not use tools."
}
],
"events": [
{
"id": "035a9501-7f67-42b7-a2f5-994df38cff15",
"createdAt": "2026-11-03T11:07:01.970Z",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"kind": "model",
"source": "model_connection",
"label": "单次 Messages 身份预检(无工具)",
"phase": "failed",
"contextPlanId": "3c21957a-6ce7-4682-a24e-0212eace4a2a",
"startedAt": "2026-11-03T11:07:01.970Z",
"modelAttemptId": "41c28fcb-4300-4005-bc2c-4b7fce3b2bf3",
"connectionId": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"actualModel": "claude-owned-m",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"diagnostic": {
"id": "42238d3b-0f8b-4256-b188-2aef8ec619ca",
"source": "model_connection",
"origin": "upstream",
"stage": "model_request",
"kind": "rate_limit",
"httpStatus": 429,
"endpoint": "http://127.0.0.1:49372",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempts": 1,
"occurredAt": "2026-11-03T11:07:02.336Z",
"retryable": true,
"title": "模型服务请求达到限额",
"upstream": {
"schema": 1,
"scope": "http-body",
"httpStatus": 429,
"code": null,
"type": "rate_limit_error",
"message": "Owned preflight rate limit",
"param": null,
"reason": null,
"requestId": "owned_m_429",
"responseId": null,
"eventType": null,
"headers": {
"content-type": "application/json",
"retry-after": "1",
"date": "Tue, 03 Nov 2026 11:07:02 GMT",
"request-id": "owned_m_429"
},
"retryAfter": "1",
"retryAfterMs": 1000,
"rawPreview": "{\"type\":\"error\",\"error\":{\"type\":\"rate_limit_error\",\"message\":\"Owned preflight rate limit\"}}",
"previewTruncated": false,
"capture": {
"scope": "http-body",
"receivedBytes": 91,
"capturedBytes": 91,
"storedBytes": 91,
"retention": "artifact",
"complete": true,
"truncated": false,
"interrupted": false,
"capturedRange": [
0,
91
],
"redacted": false,
"sha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3"
},
"artifactId": "61498b0d-0814-4aed-9ce0-c32433221d98",
"artifactSha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3",
"artifactScope": "http-body",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40"
},
"upstreamRequestId": "owned_m_429",
"code": null,
"type": "rate_limit_error",
"retryAfterMs": 1000,
"limitReason": "rate",
"wireAttemptId": "a4805e41-b1b1-4aaf-9533-8b230f3878ae",
"message": "模型服务请求达到限额(HTTP 429)。 Owned preflight rate limit",
"detail": "Owned preflight rate limit",
"suggestion": "依据上游错误码核对限制原因;有 Retry-After 时按原等待时间处理,原因未给出时保持未确定。",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"responseUrl": "http://127.0.0.1:49372/v1/messages",
"redirected": false,
"redirectMode": "manual",
"retryDecision": "attempt_limit",
"remainingWaitBudgetMs": 299634
},
"finishedAt": "2026-11-03T11:07:02.655Z"
}
],
"changes": [],
"approvals": [],
"failure": {
"id": "42238d3b-0f8b-4256-b188-2aef8ec619ca",
"source": "model_connection",
"origin": "upstream",
"stage": "model_request",
"kind": "rate_limit",
"httpStatus": 429,
"endpoint": "http://127.0.0.1:49372",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempts": 1,
"occurredAt": "2026-11-03T11:07:02.336Z",
"retryable": true,
"title": "模型服务请求达到限额",
"upstream": {
"schema": 1,
"scope": "http-body",
"httpStatus": 429,
"code": null,
"type": "rate_limit_error",
"message": "Owned preflight rate limit",
"param": null,
"reason": null,
"requestId": "owned_m_429",
"responseId": null,
"eventType": null,
"headers": {
"content-type": "application/json",
"retry-after": "1",
"date": "Tue, 03 Nov 2026 11:07:02 GMT",
"request-id": "owned_m_429"
},
"retryAfter": "1",
"retryAfterMs": 1000,
"rawPreview": "{\"type\":\"error\",\"error\":{\"type\":\"rate_limit_error\",\"message\":\"Owned preflight rate limit\"}}",
"previewTruncated": false,
"capture": {
"scope": "http-body",
"receivedBytes": 91,
"capturedBytes": 91,
"storedBytes": 91,
"retention": "artifact",
"complete": true,
"truncated": false,
"interrupted": false,
"capturedRange": [
0,
91
],
"redacted": false,
"sha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3"
},
"artifactId": "61498b0d-0814-4aed-9ce0-c32433221d98",
"artifactSha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3",
"artifactScope": "http-body",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40"
},
"upstreamRequestId": "owned_m_429",
"code": null,
"type": "rate_limit_error",
"retryAfterMs": 1000,
"limitReason": "rate",
"wireAttemptId": "a4805e41-b1b1-4aaf-9533-8b230f3878ae",
"message": "模型服务请求达到限额(HTTP 429)。 Owned preflight rate limit",
"detail": "Owned preflight rate limit",
"suggestion": "依据上游错误码核对限制原因;有 Retry-After 时按原等待时间处理,原因未给出时保持未确定。",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"responseUrl": "http://127.0.0.1:49372/v1/messages",
"redirected": false,
"redirectMode": "manual",
"retryDecision": "attempt_limit",
"remainingWaitBudgetMs": 299634,
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"messageId": "fbbe79d4-fa56-4a78-9f84-c5d53d50be76"
},
"lastFailure": {
"id": "42238d3b-0f8b-4256-b188-2aef8ec619ca",
"source": "model_connection",
"origin": "upstream",
"stage": "model_request",
"kind": "rate_limit",
"httpStatus": 429,
"endpoint": "http://127.0.0.1:49372",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempts": 1,
"occurredAt": "2026-11-03T11:07:02.336Z",
"retryable": true,
"title": "模型服务请求达到限额",
"upstream": {
"schema": 1,
"scope": "http-body",
"httpStatus": 429,
"code": null,
"type": "rate_limit_error",
"message": "Owned preflight rate limit",
"param": null,
"reason": null,
"requestId": "owned_m_429",
"responseId": null,
"eventType": null,
"headers": {
"content-type": "application/json",
"retry-after": "1",
"date": "Tue, 03 Nov 2026 11:07:02 GMT",
"request-id": "owned_m_429"
},
"retryAfter": "1",
"retryAfterMs": 1000,
"rawPreview": "{\"type\":\"error\",\"error\":{\"type\":\"rate_limit_error\",\"message\":\"Owned preflight rate limit\"}}",
"previewTruncated": false,
"capture": {
"scope": "http-body",
"receivedBytes": 91,
"capturedBytes": 91,
"storedBytes": 91,
"retention": "artifact",
"complete": true,
"truncated": false,
"interrupted": false,
"capturedRange": [
0,
91
],
"redacted": false,
"sha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3"
},
"artifactId": "61498b0d-0814-4aed-9ce0-c32433221d98",
"artifactSha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3",
"artifactScope": "http-body",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40"
},
"upstreamRequestId": "owned_m_429",
"code": null,
"type": "rate_limit_error",
"retryAfterMs": 1000,
"limitReason": "rate",
"wireAttemptId": "a4805e41-b1b1-4aaf-9533-8b230f3878ae",
"message": "模型服务请求达到限额(HTTP 429)。 Owned preflight rate limit",
"detail": "Owned preflight rate limit",
"suggestion": "依据上游错误码核对限制原因;有 Retry-After 时按原等待时间处理,原因未给出时保持未确定。",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"responseUrl": "http://127.0.0.1:49372/v1/messages",
"redirected": false,
"redirectMode": "manual",
"retryDecision": "attempt_limit",
"remainingWaitBudgetMs": 299634,
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00"
},
"compactions": 0,
"estimatedTokens": 115,
"usage": null,
"archived": false,
"revision": 17,
"eventSeq": 17,
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00"
},
"execution": {
"schemaVersion": 1,
"task": {
"schemaVersion": 1,
"id": "88676663-903a-4802-ad8c-375d53b2ff16",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"goal": "Reply OK owned-m. Do not use tools.",
"instructions": [],
"constraints": [],
"acceptance": [],
"status": "failed",
"previousTaskId": null,
"createdAt": "2026-11-03T11:07:01.580Z",
"updatedAt": "2026-11-03T11:07:02.719Z",
"activeRunId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"usage": {
"requests": 1,
"inputTokens": 0,
"outputTokens": 0,
"totalTokens": 0,
"unknownRequests": 1,
"estimatedUnknownTokens": 371
},
"budget": {
"maxTokens": 16000
},
"verification": {
"status": "not_configured",
"checkJobId": null
},
"origin": "workbench",
"contract": {
"schemaVersion": 1,
"goal": "Reply OK owned-m. Do not use tools.",
"kind": "general",
"checks": "unspecified"
},
"workspaceIdentity": {
"kind": "local-project",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\429\\project"
},
"stagnation": null
},
"run": {
"schemaVersion": 1,
"id": "11b42814-5702-46ff-9d4e-b6e532053d00",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"engine": "workbench",
"status": "failed",
"startedAt": "2026-11-03T11:07:01.580Z",
"finishedAt": "2026-11-03T11:07:02.719Z",
"resumeOf": null,
"model": {
"mode": "compatible",
"baseUrl": "http://127.0.0.1:49372",
"model": "claude-owned-m",
"contextLimit": 16000,
"maxOutputTokens": 256,
"configRevision": 1,
"modelResponseTimeoutMs": 300000,
"modelStreamIdleTimeoutMs": 120000,
"protocol": "anthropic-messages",
"streaming": true,
"messagesAuth": "bearer",
"nativeParameters": {},
"outputTokenField": "max_tokens",
"routing": {
"type": "connection",
"id": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"revision": 1,
"strategy": "round-robin",
"sticky": true,
"maxAttempts": 1,
"queueTimeoutMs": 15000,
"members": [
{
"id": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"name": "owned M 429",
"brand": "自定义",
"protocol": "anthropic-messages",
"model": "claude-owned-m",
"capabilitySchema": 1,
"capabilityDeclarations": {
"text": "supported",
"tools": "supported",
"streaming": "supported",
"vision": "unsupported",
"audio": "unknown"
},
"streaming": true,
"nativeParameters": {},
"outputTokenField": "max_tokens",
"modelIdentityPolicy": "require-match",
"modelIdentityAliases": [],
"modelIdentityAliasSource": "",
"messagesAuth": "bearer",
"baseUrl": "http://127.0.0.1:49372",
"enabled": true,
"requiresKey": true,
"contextLimit": 16000,
"maxOutputTokens": 256,
"concurrency": 1,
"revision": 1,
"source": "manual",
"createdAt": "2026-11-03T11:07:01.447Z",
"updatedAt": "2026-11-03T11:07:01.447Z",
"weight": 1,
"credentialRevision": 1
}
]
}
},
"budget": {
"maxSteps": 1,
"maxDurationMs": 320000,
"maxTokens": 16000,
"maxTaskTokens": 16000
},
"usage": {
"requests": 1,
"inputTokens": 0,
"outputTokens": 0,
"totalTokens": 0,
"unknownRequests": 1,
"estimatedUnknownTokens": 371
},
"steps": 1,
"toolCalls": 0,
"stopReason": "rate_limit",
"checkpoint": {
"stepId": "7caa32dc-d74a-4920-b5d0-e476cdb99d63",
"completedToolIds": [],
"pendingToolIds": [],
"requiresVerification": []
},
"origin": "workbench",
"purpose": "execution",
"workspaceIdentity": {
"kind": "local-project",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\429\\project"
},
"currentContextPlanId": "3c21957a-6ce7-4682-a24e-0212eace4a2a",
"consumption": [
{
"attemptId": "41c28fcb-4300-4005-bc2c-4b7fce3b2bf3",
"requestId": "3c21957a-6ce7-4682-a24e-0212eace4a2a",
"connectionId": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"connectionName": "owned M 429",
"model": "claude-owned-m",
"endpoint": "http://127.0.0.1:49372",
"protocol": "anthropic-messages",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"responseUrl": "http://127.0.0.1:49372/v1/messages",
"redirected": false,
"redirectMode": "manual",
"startedAt": "2026-11-03T11:07:01.970Z",
"finishedAt": "2026-11-03T11:07:02.403Z",
"outcome": "failed",
"usageStatus": "unknown",
"usage": null,
"localEstimate": {
"schemaVersion": 1,
"inputTokens": 115,
"outputTokens": 256,
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"requestBytes": 320,
"estimateMethod": "utf8-third-plus-framing-v1"
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [],
"status": "incomplete",
"complete": false,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights."
},
"estimatedUnknownTokens": 371
}
]
},
"workspaceIdentity": {
"kind": "local-project",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\429\\project"
},
"workspaces": [],
"runs": [
{
"id": "11b42814-5702-46ff-9d4e-b6e532053d00",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"status": "failed",
"startedAt": "2026-11-03T11:07:01.580Z",
"finishedAt": "2026-11-03T11:07:02.719Z",
"steps": 1,
"toolCalls": 0,
"usage": {
"requests": 1,
"inputTokens": 0,
"outputTokens": 0,
"totalTokens": 0,
"unknownRequests": 1,
"estimatedUnknownTokens": 371
},
"stopReason": "rate_limit",
"model": {
"mode": "compatible",
"baseUrl": "http://127.0.0.1:49372",
"model": "claude-owned-m",
"contextLimit": 16000,
"maxOutputTokens": 256,
"configRevision": 1,
"modelResponseTimeoutMs": 300000,
"modelStreamIdleTimeoutMs": 120000,
"protocol": "anthropic-messages",
"streaming": true,
"messagesAuth": "bearer",
"nativeParameters": {},
"outputTokenField": "max_tokens",
"routing": {
"type": "connection",
"id": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"revision": 1,
"strategy": "round-robin",
"sticky": true,
"maxAttempts": 1,
"queueTimeoutMs": 15000,
"members": [
{
"id": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"name": "owned M 429",
"brand": "自定义",
"protocol": "anthropic-messages",
"model": "claude-owned-m",
"capabilitySchema": 1,
"capabilityDeclarations": {
"text": "supported",
"tools": "supported",
"streaming": "supported",
"vision": "unsupported",
"audio": "unknown"
},
"streaming": true,
"nativeParameters": {},
"outputTokenField": "max_tokens",
"modelIdentityPolicy": "require-match",
"modelIdentityAliases": [],
"modelIdentityAliasSource": "",
"messagesAuth": "bearer",
"baseUrl": "http://127.0.0.1:49372",
"enabled": true,
"requiresKey": true,
"contextLimit": 16000,
"maxOutputTokens": 256,
"concurrency": 1,
"revision": 1,
"source": "manual",
"createdAt": "2026-11-03T11:07:01.447Z",
"updatedAt": "2026-11-03T11:07:01.447Z",
"weight": 1,
"credentialRevision": 1
}
]
}
},
"budget": {
"maxSteps": 1,
"maxDurationMs": 320000,
"maxTokens": 16000,
"maxTaskTokens": 16000
},
"resumeOf": null,
"purpose": "execution",
"consumption": [
{
"attemptId": "41c28fcb-4300-4005-bc2c-4b7fce3b2bf3",
"requestId": "3c21957a-6ce7-4682-a24e-0212eace4a2a",
"connectionId": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"connectionName": "owned M 429",
"model": "claude-owned-m",
"endpoint": "http://127.0.0.1:49372",
"protocol": "anthropic-messages",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"responseUrl": "http://127.0.0.1:49372/v1/messages",
"redirected": false,
"redirectMode": "manual",
"startedAt": "2026-11-03T11:07:01.970Z",
"finishedAt": "2026-11-03T11:07:02.403Z",
"outcome": "failed",
"usageStatus": "unknown",
"usage": null,
"localEstimate": {
"schemaVersion": 1,
"inputTokens": 115,
"outputTokens": 256,
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"requestBytes": 320,
"estimateMethod": "utf8-third-plus-framing-v1"
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [],
"status": "incomplete",
"complete": false,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights."
},
"estimatedUnknownTokens": 371
}
],
"workspaceIdentity": {
"kind": "local-project",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\429\\project"
}
}
],
"progress": {
"totalTools": 0,
"completedTools": 0,
"requiresVerification": []
},
"artifacts": [
{
"id": "cf270e63-13cc-43d2-9863-92af0bf95143",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"kind": "model-request",
"mediaType": "application/json",
"sha256": "62c4ff4737adb65b4bf86cbd43ba09adc879ef18ac678f3d74d095120471202a",
"bytes": 699,
"relative": "62/62c4ff4737adb65b4bf86cbd43ba09adc879ef18ac678f3d74d095120471202a.blob",
"createdAt": "2026-11-03T11:07:01.762Z",
"status": "stored"
},
{
"id": "4a0b9ea9-1e33-4c18-ad43-a2905b47d096",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"kind": "context-plan",
"mediaType": "application/json",
"sha256": "2e8d7af535b361989f1df2b31f541bf78c8a69c5b3bf94852b2fa6e2ebc31109",
"bytes": 2083,
"relative": "2e/2e8d7af535b361989f1df2b31f541bf78c8a69c5b3bf94852b2fa6e2ebc31109.blob",
"createdAt": "2026-11-03T11:07:01.795Z",
"status": "stored"
},
{
"id": "03055e60-b6c0-47e9-ac4b-60a2436989c3",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"kind": "model-attempt-request",
"mediaType": "application/json",
"sha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"bytes": 320,
"relative": "34/341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85.blob",
"createdAt": "2026-11-03T11:07:01.870Z",
"status": "stored"
},
{
"id": "61498b0d-0814-4aed-9ce0-c32433221d98",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"kind": "model-http-response",
"mediaType": "text/plain",
"sha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3",
"bytes": 91,
"relative": "29/29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3.blob",
"createdAt": "2026-11-03T11:07:02.270Z",
"status": "stored"
}
],
"checks": [],
"browsingTabs": [],
"browsingActions": [],
"previewPlans": [],
"previewJobs": [],
"mcpJobs": [],
"taskPlans": [],
"taskProofs": [],
"researchTargets": [],
"researchJobs": [],
"engineJobs": [],
"browserPlans": [],
"browserJobs": [],
"contextPlans": [
{
"schemaVersion": 1,
"id": "3c21957a-6ce7-4682-a24e-0212eace4a2a",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"stepId": "7caa32dc-d74a-4920-b5d0-e476cdb99d63",
"requestIndex": 1,
"status": "failed",
"createdAt": "2026-11-03T11:07:01.762Z",
"finishedAt": "2026-11-03T11:07:02.594Z",
"requestSha256": "62c4ff4737adb65b4bf86cbd43ba09adc879ef18ac678f3d74d095120471202a",
"requestBytes": 699,
"estimatedInputTokens": 115,
"requestArtifactId": "cf270e63-13cc-43d2-9863-92af0bf95143",
"tokenReservation": {
"schemaVersion": 1,
"method": "observed-route-max-v1",
"overheadTokens": 0,
"reservedTokens": 371,
"routes": [
{
"connectionId": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"model": "claude-owned-m",
"protocol": "anthropic-messages",
"endpoint": "http://127.0.0.1:49372",
"localInputTokens": 115,
"outputReservation": 256,
"inputRatio": 1,
"observedOutputTokens": 256,
"observedTotalFloor": 0,
"reservedTokens": 371,
"sources": []
}
]
},
"maxBudgetAttempts": 8,
"estimateMethod": "maximum-wire-variant-utf8-v1",
"beforeInputTokens": 115,
"beforeBytes": 699,
"budget": {
"contextLimit": 16000,
"outputReservation": 256,
"safetyTokens": 1600,
"inputBudget": 14144,
"targetInputTokens": 14144,
"runRemainingTokens": 16000,
"taskRemainingTokens": 16000,
"limitingReason": "context_limit"
},
"mandatoryInputTokens": 115,
"nextReadFixedInputTokens": 115,
"workingRetentionTokens": 0,
"workingRetentionPolicyVersion": 1,
"originalMessages": 1,
"retainedMessages": 1,
"archivedGroups": 0,
"retainedGroups": [
0
],
"historyIndexArtifactId": null,
"historyIndexSha256": null,
"actualUsage": null,
"workingSetArtifactId": null,
"workingSetSha256": null,
"workingSetEntries": 0,
"workingSetViewVersion": 2,
"latestToolsArtifactId": null,
"latestToolsSha256": null,
"latestToolsEntries": 0,
"projectMemoryArtifactId": null,
"projectMemorySha256": null,
"projectMemoryEntries": 0,
"skillCatalogArtifactId": null,
"skillCatalogSha256": null,
"skillCatalogEntries": 0,
"mcpCatalogArtifactId": null,
"mcpCatalogSha256": null,
"mcpCatalogEntries": 0,
"requestAttempts": 1,
"responseSha256": null,
"source": "workbench",
"artifactId": "4a0b9ea9-1e33-4c18-ad43-a2905b47d096",
"error": "模型服务请求达到限额(HTTP 429)。 Owned preflight rate limit"
}
],
"changeSets": [],
"plans": [],
"deliveries": [],
"deliveryPlans": [],
"accountedTaskTokens": 371
}
}
@@ -0,0 +1,83 @@
(async function nativeMessagesProbe(input) {
const ensure = (value, message) => { if (!value) throw new Error(message); };
ensure(process.pid === input.pid, 'Unexpected host PID.');
const servers = process._getActiveHandles().filter(h => h?.listening && h.address?.()?.address === '127.0.0.1' && h.address?.()?.port === input.port);
ensure(servers.length === 1, 'Expected one owned loopback server.');
const context = servers[0].workbenchHostAuthority?.context;
ensure(context?.provider && context?.store && context?.runtime, 'Original context missing.');
const { store, provider, runtime } = context;
ensure(runtime.appDirectory === input.runtimeRoot, 'Loaded runtime changed.');
ensure(![...store.sessions.values()].some(s => ['running', 'stopping', 'awaiting_approval'].includes(s.status)), 'Host is busy.');
const session = store.session(input.sessionId), project = store.project(session.projectId);
ensure(session.projectId === input.projectId && session.messages.length === 0 && !session.runId, 'Probe session must be new and owned.');
const hub = provider.modelHub(store);
const connection = hub.view().connections.find(c => c.id === input.connectionId);
ensure(connection?.hasKey && connection.credentialRevision === input.credentialRevision, 'Original saved credential revision changed.');
ensure(connection.protocol === 'anthropic-messages' && connection.messagesAuth === 'bearer' && connection.modelIdentityPolicy === 'require-match', 'Probe protocol/auth/identity mismatch.');
ensure(connection.baseUrl === input.baseUrl && connection.model === input.model && connection.revision === input.connectionRevision, 'Connection changed.');
const config = hub.prepare({ ...store.config, mode: 'compatible', maxSteps: 1, maxOutputTokens: 256,
contextLimit: 16000, maxRunTokens: 16000, maxTaskTokens: 16000, maxRunDurationMs: 320000,
modelResponseTimeoutMs: 300000, modelStreamIdleTimeoutMs: 120000 }, { type: 'connection', id: input.connectionId });
ensure(config.modelRouting.members.length === 1 && config.modelRouting.maxAttempts === 1 && config.maxOutputTokens === 256, 'One-request hard limits missing.');
globalThis.__messagesMProbeOnce ??= new Set();
ensure(!globalThis.__messagesMProbeOnce.has(input.sessionId), 'This sample cannot be replayed.');
globalThis.__messagesMProbeOnce.add(input.sessionId);
session.title = input.title;
const run = store.beginRun(session, input.prompt, false, config, { forceNewTask: true });
store.message(session, 'user', input.prompt);
session.providerMessages.push({ role: 'user', content: input.prompt });
const step = store.beginStep(session, run), planner = runtime.requestPlanner();
let plan, response, diagnostic = null, requestEvent, receipts = 0, wireStarts = 0;
const wireIds = new Set();
const signal = AbortSignal.timeout(320000);
let stopReason = 'identity_probe_complete';
try {
plan = planner.prepare(session, run, config, { system: 'This is a connection identity preflight. Reply to the user exactly. No tools are available. Do not perform any project work.', tools: [], stepId: step.id });
const variants = plan.body.kind === 'model-route-plan' ? plan.body.variants.map(v => v.body) : [plan.body];
ensure(variants.length === 1 && variants.every(b => !(b.tools?.length) && b.max_tokens === 256 && b.model === input.model), 'Wire request violates no-tool contract.');
ensure(plan.record.tokenReservation.reservedTokens <= 16000, 'Reservation exceeds sealed budget.');
const progress = event => {
if (event.consumption) {
receipts += Number(store.recordConsumption(session, run, { ...event.consumption, requestId: plan.record.id }, plan.record.tokenReservation.reservedTokens));
return;
}
if (event.phase === 'running' && event.wireAttemptId) { wireIds.add(event.wireAttemptId); wireStarts = wireIds.size; ensure(wireStarts === 1, 'No retries permitted.'); }
requestEvent ??= store.event(session, { kind: 'model', source: 'model_connection', label: '单次 Messages 身份预检(无工具)', phase: 'running', runId: run.id, contextPlanId: plan.record.id, startedAt: new Date().toISOString() });
if (event.modelAttemptId) requestEvent.modelAttemptId = event.modelAttemptId;
if (event.connectionId) requestEvent.connectionId = event.connectionId;
if (event.model) requestEvent.actualModel = event.model;
if (event.modelIdentity) requestEvent.modelIdentity = event.modelIdentity;
if (event.requestUrl) requestEvent.requestUrl = event.requestUrl;
if (event.diagnostic) requestEvent.diagnostic = event.diagnostic;
if (event.stream) requestEvent.stream = event.stream;
if (event.phase) requestEvent.phase = event.phase;
plan.record.requestAttempts = wireStarts; step.modelAttempts = wireStarts;
store.save(session);
};
response = await provider.complete(config, plan.messages, project, signal, progress, plan.record, [], { attempts: 1, maxBudgetAttempts: 1 });
ensure(!response.message.tool_calls?.length, 'Unsolicited tools are never executed.');
ensure(receipts === 1, 'Expected one original consumption receipt.');
planner.finish(session, plan.record, { response, attempts: wireStarts });
session.providerMessages.push(response.message); store.message(session, 'assistant', response.message.content);
const total = (run.usage?.reportedTokens ?? run.usage?.totalTokens ?? 0);
session.status = total > 16000 ? 'paused' : 'completed';
if (total > 16000) stopReason = 'task_token_limit';
} catch (cause) {
diagnostic = cause.diagnostic ?? { source: 'workbench', origin: 'local', stage: 'identity_preflight', kind: signal.aborted ? 'cancelled' : 'runtime', title: '身份预检需要核对', message: provider.redact(cause.message), occurredAt: new Date().toISOString() };
session.status = diagnostic.kind === 'model_identity' || signal.aborted ? 'paused' : 'failed';
stopReason = diagnostic.kind;
session.lastFailure = { ...diagnostic, runId: run.id };
const message = store.message(session, 'system', diagnostic.message);
session.failure = { ...session.lastFailure, messageId: message.id };
if (plan) planner.finish(session, plan.record, { error: new Error(diagnostic.message), attempts: wireStarts });
} finally {
if (requestEvent) { requestEvent.phase = diagnostic ? 'failed' : 'completed'; requestEvent.finishedAt = new Date().toISOString(); if (diagnostic) requestEvent.diagnostic = diagnostic; }
store.finishStep(session, step, diagnostic ? 'failed' : 'completed');
store.finishRun(session, run, stopReason);
}
const attempts = hub.attempts.filter(a => a.sessionId === session.id);
ensure(attempts.length <= 1 && wireStarts <= 1, 'Single-request limit exceeded.');
return { sessionId: session.id, projectId: project.id, taskId: run.taskId, runId: run.id, status: session.status,
stopReason, attempts: attempts.length, receipts, wireStarts, actualTools: run.toolCalls, usage: run.usage,
contextPlanId: plan?.record.id ?? null, diagnostic, message: response?.message ?? null };
})({"pid":12072,"port":49377,"sessionId":"0bedf6de-e5c5-4a1d-8125-cfd4acf96d40","projectId":"29fbf823-c158-4131-8d7d-bd8fd5d204d4","runtimeRoot":"D:\\Ai-project\\REA逆向\\Agent-Workbench","connectionId":"9386719f-b632-4fe8-bd0e-7f9f31ca83f4","credentialRevision":1,"connectionRevision":1,"baseUrl":"http://127.0.0.1:49372","model":"claude-owned-m","title":"Owned M probe 429","prompt":"Reply OK owned-m. Do not use tools."})
@@ -0,0 +1,14 @@
{
"pid": 12072,
"port": 49377,
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"runtimeRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench",
"connectionId": "9386719f-b632-4fe8-bd0e-7f9f31ca83f4",
"credentialRevision": 1,
"connectionRevision": 1,
"baseUrl": "http://127.0.0.1:49372",
"model": "claude-owned-m",
"title": "Owned M probe 429",
"prompt": "Reply OK owned-m. Do not use tools."
}
@@ -0,0 +1,95 @@
{
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40",
"projectId": "29fbf823-c158-4131-8d7d-bd8fd5d204d4",
"taskId": "88676663-903a-4802-ad8c-375d53b2ff16",
"runId": "11b42814-5702-46ff-9d4e-b6e532053d00",
"status": "failed",
"stopReason": "rate_limit",
"attempts": 1,
"receipts": 1,
"wireStarts": 1,
"actualTools": 0,
"usage": {
"requests": 1,
"inputTokens": 0,
"outputTokens": 0,
"totalTokens": 0,
"unknownRequests": 1,
"estimatedUnknownTokens": 371
},
"contextPlanId": "3c21957a-6ce7-4682-a24e-0212eace4a2a",
"diagnostic": {
"id": "42238d3b-0f8b-4256-b188-2aef8ec619ca",
"source": "model_connection",
"origin": "upstream",
"stage": "model_request",
"kind": "rate_limit",
"httpStatus": 429,
"endpoint": "http://127.0.0.1:49372",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempts": 1,
"occurredAt": "2026-11-03T11:07:02.336Z",
"retryable": true,
"title": "模型服务请求达到限额",
"upstream": {
"schema": 1,
"scope": "http-body",
"httpStatus": 429,
"code": null,
"type": "rate_limit_error",
"message": "Owned preflight rate limit",
"param": null,
"reason": null,
"requestId": "owned_m_429",
"responseId": null,
"eventType": null,
"headers": {
"content-type": "application/json",
"retry-after": "1",
"date": "Tue, 03 Nov 2026 11:07:02 GMT",
"request-id": "owned_m_429"
},
"retryAfter": "1",
"retryAfterMs": 1000,
"rawPreview": "{\"type\":\"error\",\"error\":{\"type\":\"rate_limit_error\",\"message\":\"Owned preflight rate limit\"}}",
"previewTruncated": false,
"capture": {
"scope": "http-body",
"receivedBytes": 91,
"capturedBytes": 91,
"storedBytes": 91,
"retention": "artifact",
"complete": true,
"truncated": false,
"interrupted": false,
"capturedRange": [
0,
91
],
"redacted": false,
"sha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3"
},
"artifactId": "61498b0d-0814-4aed-9ce0-c32433221d98",
"artifactSha256": "29e99cbff0239027f6714443849939800983f9f587bd08a73b283996a667baa3",
"artifactScope": "http-body",
"sessionId": "0bedf6de-e5c5-4a1d-8125-cfd4acf96d40"
},
"upstreamRequestId": "owned_m_429",
"code": null,
"type": "rate_limit_error",
"retryAfterMs": 1000,
"limitReason": "rate",
"wireAttemptId": "a4805e41-b1b1-4aaf-9533-8b230f3878ae",
"message": "模型服务请求达到限额(HTTP 429)。 Owned preflight rate limit",
"detail": "Owned preflight rate limit",
"suggestion": "依据上游错误码核对限制原因;有 Retry-After 时按原等待时间处理,原因未给出时保持未确定。",
"requestUrl": "http://127.0.0.1:49372/v1/messages",
"responseUrl": "http://127.0.0.1:49372/v1/messages",
"redirected": false,
"redirectMode": "manual",
"retryDecision": "attempt_limit",
"remainingWaitBudgetMs": 299634
},
"message": null
}
@@ -0,0 +1,29 @@
[
{
"route": "/v1/messages",
"body": {
"model": "claude-owned-m",
"system": [
{
"type": "text",
"text": "This is a connection identity preflight. Reply to the user exactly. No tools are available. Do not perform any project work."
}
],
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": "Reply OK owned-m. Do not use tools."
}
]
}
],
"max_tokens": 256,
"stream": true
},
"authCorrect": true,
"version": "2023-06-01"
}
]
@@ -0,0 +1,94 @@
{
"id": "9b8da8f2-13b6-486a-9456-5429bc330be4",
"requestId": "71ddf33c-340e-496a-9287-6cd71f3e9870",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"connectionId": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"connectionRevision": 1,
"credentialRevision": 1,
"poolId": null,
"poolRevision": 1,
"model": "claude-owned-m",
"protocol": "anthropic-messages",
"streaming": true,
"endpoint": "http://127.0.0.1:49285",
"requestUrl": "http://127.0.0.1:49285/v1/messages",
"messagesAuth": "bearer",
"ordinal": 1,
"status": "completed",
"startedAt": "2026-11-03T11:06:54.074Z",
"requestBytes": 320,
"requirements": [
"text"
],
"rejections": [],
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"usage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"usageStatus": "reported",
"requestArtifactId": "df54b864-b9c4-4170-b675-e9ef6a0cef3c",
"wireAttemptId": "f99f3053-cdc3-4991-88c5-6d4890a84723",
"upstreamResponse": {
"httpStatus": 200,
"requestId": "owned_m_match",
"headers": {
"content-type": "text/event-stream",
"date": "Tue, 03 Nov 2026 11:06:54 GMT",
"request-id": "owned_m_match"
},
"requestUrl": "http://127.0.0.1:49285/v1/messages",
"responseUrl": "http://127.0.0.1:49285/v1/messages",
"redirected": false,
"redirectMode": "manual"
},
"partial": {
"sequence": 6,
"bytes": 726,
"textCharacters": 10,
"textPreview": "OK owned-m",
"toolFragments": 0
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"claude-owned-m"
],
"status": "matched",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "2220fd4a-f540-435b-a6fc-1ec1b2a6bf97",
"sha256": "809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"scope": "sse-stream"
}
},
"stream": {
"bytes": 726,
"frames": 6,
"sha256": "809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe",
"terminal": true,
"protocol": "anthropic-messages"
},
"responseArtifactId": "2220fd4a-f540-435b-a6fc-1ec1b2a6bf97",
"responseSha256": "809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe",
"responseRedacted": false,
"finishedAt": "2026-11-03T11:06:54.747Z"
}
@@ -0,0 +1,673 @@
{
"schemaVersion": 1,
"executionSchemaVersion": 1,
"exportedAt": "2026-11-03T11:06:55.206Z",
"project": {
"id": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\match\\project",
"name": "project",
"addedAt": "2026-11-03T11:06:53.573Z"
},
"session": {
"id": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"title": "Owned M probe match",
"status": "completed",
"createdAt": "2026-11-03T11:06:53.600Z",
"updatedAt": "2026-11-03T11:06:54.939Z",
"messages": [
{
"id": "27e12eb2-dd4b-4cb6-b9dd-887b478c0bce",
"role": "user",
"content": "Reply OK owned-m. Do not use tools.",
"createdAt": "2026-11-03T11:06:53.864Z"
},
{
"id": "8b7489b0-967c-4cbb-b6dd-0a011cc0321a",
"role": "assistant",
"content": "OK owned-m",
"createdAt": "2026-11-03T11:06:54.830Z"
}
],
"providerMessages": [
{
"role": "user",
"content": "Reply OK owned-m. Do not use tools."
},
{
"role": "assistant",
"content": "OK owned-m",
"providerState": {
"protocol": "anthropic-messages",
"endpoint": "http://127.0.0.1:49285",
"model": "claude-owned-m",
"responseId": "msg_owned_m",
"content": [
{
"type": "text",
"text": "OK owned-m"
}
]
}
}
],
"events": [
{
"id": "e58e6037-bd65-4ace-a5ac-e5d8e7278f5e",
"createdAt": "2026-11-03T11:06:54.224Z",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"kind": "model",
"source": "model_connection",
"label": "单次 Messages 身份预检(无工具)",
"phase": "completed",
"contextPlanId": "71ddf33c-340e-496a-9287-6cd71f3e9870",
"startedAt": "2026-11-03T11:06:54.223Z",
"modelAttemptId": "9b8da8f2-13b6-486a-9456-5429bc330be4",
"connectionId": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"actualModel": "claude-owned-m",
"requestUrl": "http://127.0.0.1:49285/v1/messages",
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"claude-owned-m"
],
"status": "matched",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "2220fd4a-f540-435b-a6fc-1ec1b2a6bf97",
"sha256": "809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"scope": "sse-stream"
}
},
"stream": {
"bytes": 726,
"frames": 6,
"sha256": "809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe",
"terminal": true,
"protocol": "anthropic-messages"
},
"finishedAt": "2026-11-03T11:06:54.880Z"
}
],
"changes": [],
"approvals": [],
"failure": null,
"lastFailure": null,
"compactions": 0,
"estimatedTokens": 115,
"usage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"archived": false,
"revision": 19,
"eventSeq": 19,
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b"
},
"execution": {
"schemaVersion": 1,
"task": {
"schemaVersion": 1,
"id": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"goal": "Reply OK owned-m. Do not use tools.",
"instructions": [],
"constraints": [],
"acceptance": [],
"status": "completed",
"previousTaskId": null,
"createdAt": "2026-11-03T11:06:53.813Z",
"updatedAt": "2026-11-03T11:06:54.939Z",
"activeRunId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"usage": {
"requests": 1,
"inputTokens": 40,
"outputTokens": 5,
"totalTokens": 45,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"budget": {
"maxTokens": 16000
},
"verification": {
"status": "not_configured",
"checkJobId": null
},
"origin": "workbench",
"contract": {
"schemaVersion": 1,
"goal": "Reply OK owned-m. Do not use tools.",
"kind": "general",
"checks": "unspecified"
},
"workspaceIdentity": {
"kind": "local-project",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\match\\project"
},
"stagnation": null
},
"run": {
"schemaVersion": 1,
"id": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"engine": "workbench",
"status": "completed",
"startedAt": "2026-11-03T11:06:53.813Z",
"finishedAt": "2026-11-03T11:06:54.939Z",
"resumeOf": null,
"model": {
"mode": "compatible",
"baseUrl": "http://127.0.0.1:49285",
"model": "claude-owned-m",
"contextLimit": 16000,
"maxOutputTokens": 256,
"configRevision": 1,
"modelResponseTimeoutMs": 300000,
"modelStreamIdleTimeoutMs": 120000,
"protocol": "anthropic-messages",
"streaming": true,
"messagesAuth": "bearer",
"nativeParameters": {},
"outputTokenField": "max_tokens",
"routing": {
"type": "connection",
"id": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"revision": 1,
"strategy": "round-robin",
"sticky": true,
"maxAttempts": 1,
"queueTimeoutMs": 15000,
"members": [
{
"id": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"name": "owned M match",
"brand": "自定义",
"protocol": "anthropic-messages",
"model": "claude-owned-m",
"capabilitySchema": 1,
"capabilityDeclarations": {
"text": "supported",
"tools": "supported",
"streaming": "supported",
"vision": "unsupported",
"audio": "unknown"
},
"streaming": true,
"nativeParameters": {},
"outputTokenField": "max_tokens",
"modelIdentityPolicy": "require-match",
"modelIdentityAliases": [],
"modelIdentityAliasSource": "",
"messagesAuth": "bearer",
"baseUrl": "http://127.0.0.1:49285",
"enabled": true,
"requiresKey": true,
"contextLimit": 16000,
"maxOutputTokens": 256,
"concurrency": 1,
"revision": 1,
"source": "manual",
"createdAt": "2026-11-03T11:06:53.659Z",
"updatedAt": "2026-11-03T11:06:53.659Z",
"weight": 1,
"credentialRevision": 1
}
]
}
},
"budget": {
"maxSteps": 1,
"maxDurationMs": 320000,
"maxTokens": 16000,
"maxTaskTokens": 16000
},
"usage": {
"requests": 1,
"inputTokens": 40,
"outputTokens": 5,
"totalTokens": 45,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"steps": 1,
"toolCalls": 0,
"stopReason": "identity_probe_complete",
"checkpoint": {
"stepId": "123329b2-ca92-47ae-a474-287f130799f2",
"completedToolIds": [],
"pendingToolIds": [],
"requiresVerification": []
},
"origin": "workbench",
"purpose": "execution",
"workspaceIdentity": {
"kind": "local-project",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\match\\project"
},
"currentContextPlanId": "71ddf33c-340e-496a-9287-6cd71f3e9870",
"consumption": [
{
"attemptId": "9b8da8f2-13b6-486a-9456-5429bc330be4",
"requestId": "71ddf33c-340e-496a-9287-6cd71f3e9870",
"connectionId": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"connectionName": "owned M match",
"model": "claude-owned-m",
"endpoint": "http://127.0.0.1:49285",
"protocol": "anthropic-messages",
"requestUrl": "http://127.0.0.1:49285/v1/messages",
"responseUrl": "http://127.0.0.1:49285/v1/messages",
"redirected": false,
"redirectMode": "manual",
"startedAt": "2026-11-03T11:06:54.223Z",
"finishedAt": "2026-11-03T11:06:54.672Z",
"outcome": "completed",
"usageStatus": "reported",
"usage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"localEstimate": {
"schemaVersion": 1,
"inputTokens": 115,
"outputTokens": 256,
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"requestBytes": 320,
"estimateMethod": "utf8-third-plus-framing-v1"
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"claude-owned-m"
],
"status": "matched",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "2220fd4a-f540-435b-a6fc-1ec1b2a6bf97",
"sha256": "809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"scope": "sse-stream"
}
},
"estimatedUnknownTokens": 0
}
]
},
"workspaceIdentity": {
"kind": "local-project",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\match\\project"
},
"workspaces": [],
"runs": [
{
"id": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"status": "completed",
"startedAt": "2026-11-03T11:06:53.813Z",
"finishedAt": "2026-11-03T11:06:54.939Z",
"steps": 1,
"toolCalls": 0,
"usage": {
"requests": 1,
"inputTokens": 40,
"outputTokens": 5,
"totalTokens": 45,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"stopReason": "identity_probe_complete",
"model": {
"mode": "compatible",
"baseUrl": "http://127.0.0.1:49285",
"model": "claude-owned-m",
"contextLimit": 16000,
"maxOutputTokens": 256,
"configRevision": 1,
"modelResponseTimeoutMs": 300000,
"modelStreamIdleTimeoutMs": 120000,
"protocol": "anthropic-messages",
"streaming": true,
"messagesAuth": "bearer",
"nativeParameters": {},
"outputTokenField": "max_tokens",
"routing": {
"type": "connection",
"id": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"revision": 1,
"strategy": "round-robin",
"sticky": true,
"maxAttempts": 1,
"queueTimeoutMs": 15000,
"members": [
{
"id": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"name": "owned M match",
"brand": "自定义",
"protocol": "anthropic-messages",
"model": "claude-owned-m",
"capabilitySchema": 1,
"capabilityDeclarations": {
"text": "supported",
"tools": "supported",
"streaming": "supported",
"vision": "unsupported",
"audio": "unknown"
},
"streaming": true,
"nativeParameters": {},
"outputTokenField": "max_tokens",
"modelIdentityPolicy": "require-match",
"modelIdentityAliases": [],
"modelIdentityAliasSource": "",
"messagesAuth": "bearer",
"baseUrl": "http://127.0.0.1:49285",
"enabled": true,
"requiresKey": true,
"contextLimit": 16000,
"maxOutputTokens": 256,
"concurrency": 1,
"revision": 1,
"source": "manual",
"createdAt": "2026-11-03T11:06:53.659Z",
"updatedAt": "2026-11-03T11:06:53.659Z",
"weight": 1,
"credentialRevision": 1
}
]
}
},
"budget": {
"maxSteps": 1,
"maxDurationMs": 320000,
"maxTokens": 16000,
"maxTaskTokens": 16000
},
"resumeOf": null,
"purpose": "execution",
"consumption": [
{
"attemptId": "9b8da8f2-13b6-486a-9456-5429bc330be4",
"requestId": "71ddf33c-340e-496a-9287-6cd71f3e9870",
"connectionId": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"connectionName": "owned M match",
"model": "claude-owned-m",
"endpoint": "http://127.0.0.1:49285",
"protocol": "anthropic-messages",
"requestUrl": "http://127.0.0.1:49285/v1/messages",
"responseUrl": "http://127.0.0.1:49285/v1/messages",
"redirected": false,
"redirectMode": "manual",
"startedAt": "2026-11-03T11:06:54.223Z",
"finishedAt": "2026-11-03T11:06:54.672Z",
"outcome": "completed",
"usageStatus": "reported",
"usage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"localEstimate": {
"schemaVersion": 1,
"inputTokens": 115,
"outputTokens": 256,
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"requestBytes": 320,
"estimateMethod": "utf8-third-plus-framing-v1"
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"claude-owned-m"
],
"status": "matched",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "2220fd4a-f540-435b-a6fc-1ec1b2a6bf97",
"sha256": "809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"scope": "sse-stream"
}
},
"estimatedUnknownTokens": 0
}
],
"workspaceIdentity": {
"kind": "local-project",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\match\\project"
}
}
],
"progress": {
"totalTools": 0,
"completedTools": 0,
"requiresVerification": []
},
"artifacts": [
{
"id": "a3433762-d341-434e-a0a1-0bd859cea8c0",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"kind": "model-request",
"mediaType": "application/json",
"sha256": "fb88c30730a40666ec3b71d9aec409b489bac9e0cb1613394810c37dee2f6969",
"bytes": 699,
"relative": "fb/fb88c30730a40666ec3b71d9aec409b489bac9e0cb1613394810c37dee2f6969.blob",
"createdAt": "2026-11-03T11:06:53.988Z",
"status": "stored"
},
{
"id": "5c52c45f-da2c-4b61-8370-92c54fecc1d7",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"kind": "context-plan",
"mediaType": "application/json",
"sha256": "bc12b8123e495a92d877316feabd0b542d68e0fd6764901b4aa4858a6e40a123",
"bytes": 2083,
"relative": "bc/bc12b8123e495a92d877316feabd0b542d68e0fd6764901b4aa4858a6e40a123.blob",
"createdAt": "2026-11-03T11:06:54.027Z",
"status": "stored"
},
{
"id": "df54b864-b9c4-4170-b675-e9ef6a0cef3c",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"kind": "model-attempt-request",
"mediaType": "application/json",
"sha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"bytes": 320,
"relative": "34/341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85.blob",
"createdAt": "2026-11-03T11:06:54.123Z",
"status": "stored"
},
{
"id": "2220fd4a-f540-435b-a6fc-1ec1b2a6bf97",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"kind": "model-stream-response",
"mediaType": "text/event-stream",
"sha256": "809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe",
"bytes": 726,
"relative": "80/809efd6fd5d157e126bef3352440190ca23fa446f4dcbf143d235355a7bb7afe.blob",
"createdAt": "2026-11-03T11:06:54.572Z",
"status": "stored"
}
],
"checks": [],
"browsingTabs": [],
"browsingActions": [],
"previewPlans": [],
"previewJobs": [],
"mcpJobs": [],
"taskPlans": [],
"taskProofs": [],
"researchTargets": [],
"researchJobs": [],
"engineJobs": [],
"browserPlans": [],
"browserJobs": [],
"contextPlans": [
{
"schemaVersion": 1,
"id": "71ddf33c-340e-496a-9287-6cd71f3e9870",
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"stepId": "123329b2-ca92-47ae-a474-287f130799f2",
"requestIndex": 1,
"status": "completed",
"createdAt": "2026-11-03T11:06:53.988Z",
"finishedAt": "2026-11-03T11:06:54.780Z",
"requestSha256": "fb88c30730a40666ec3b71d9aec409b489bac9e0cb1613394810c37dee2f6969",
"requestBytes": 699,
"estimatedInputTokens": 115,
"requestArtifactId": "a3433762-d341-434e-a0a1-0bd859cea8c0",
"tokenReservation": {
"schemaVersion": 1,
"method": "observed-route-max-v1",
"overheadTokens": 0,
"reservedTokens": 371,
"routes": [
{
"connectionId": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"model": "claude-owned-m",
"protocol": "anthropic-messages",
"endpoint": "http://127.0.0.1:49285",
"localInputTokens": 115,
"outputReservation": 256,
"inputRatio": 1,
"observedOutputTokens": 256,
"observedTotalFloor": 0,
"reservedTokens": 371,
"sources": []
}
]
},
"maxBudgetAttempts": 8,
"estimateMethod": "maximum-wire-variant-utf8-v1",
"beforeInputTokens": 115,
"beforeBytes": 699,
"budget": {
"contextLimit": 16000,
"outputReservation": 256,
"safetyTokens": 1600,
"inputBudget": 14144,
"targetInputTokens": 14144,
"runRemainingTokens": 16000,
"taskRemainingTokens": 16000,
"limitingReason": "context_limit"
},
"mandatoryInputTokens": 115,
"nextReadFixedInputTokens": 115,
"workingRetentionTokens": 0,
"workingRetentionPolicyVersion": 1,
"originalMessages": 1,
"retainedMessages": 1,
"archivedGroups": 0,
"retainedGroups": [
0
],
"historyIndexArtifactId": null,
"historyIndexSha256": null,
"actualUsage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"workingSetArtifactId": null,
"workingSetSha256": null,
"workingSetEntries": 0,
"workingSetViewVersion": 2,
"latestToolsArtifactId": null,
"latestToolsSha256": null,
"latestToolsEntries": 0,
"projectMemoryArtifactId": null,
"projectMemorySha256": null,
"projectMemoryEntries": 0,
"skillCatalogArtifactId": null,
"skillCatalogSha256": null,
"skillCatalogEntries": 0,
"mcpCatalogArtifactId": null,
"mcpCatalogSha256": null,
"mcpCatalogEntries": 0,
"requestAttempts": 1,
"responseSha256": "3fdef2491cc795a5ff321eb481ecc86d02ec9dbde4b3b6d9ea8354cdd50b2252",
"source": "workbench",
"artifactId": "5c52c45f-da2c-4b61-8370-92c54fecc1d7"
}
],
"changeSets": [],
"plans": [],
"deliveries": [],
"deliveryPlans": [],
"accountedTaskTokens": 45
}
}
@@ -0,0 +1,83 @@
(async function nativeMessagesProbe(input) {
const ensure = (value, message) => { if (!value) throw new Error(message); };
ensure(process.pid === input.pid, 'Unexpected host PID.');
const servers = process._getActiveHandles().filter(h => h?.listening && h.address?.()?.address === '127.0.0.1' && h.address?.()?.port === input.port);
ensure(servers.length === 1, 'Expected one owned loopback server.');
const context = servers[0].workbenchHostAuthority?.context;
ensure(context?.provider && context?.store && context?.runtime, 'Original context missing.');
const { store, provider, runtime } = context;
ensure(runtime.appDirectory === input.runtimeRoot, 'Loaded runtime changed.');
ensure(![...store.sessions.values()].some(s => ['running', 'stopping', 'awaiting_approval'].includes(s.status)), 'Host is busy.');
const session = store.session(input.sessionId), project = store.project(session.projectId);
ensure(session.projectId === input.projectId && session.messages.length === 0 && !session.runId, 'Probe session must be new and owned.');
const hub = provider.modelHub(store);
const connection = hub.view().connections.find(c => c.id === input.connectionId);
ensure(connection?.hasKey && connection.credentialRevision === input.credentialRevision, 'Original saved credential revision changed.');
ensure(connection.protocol === 'anthropic-messages' && connection.messagesAuth === 'bearer' && connection.modelIdentityPolicy === 'require-match', 'Probe protocol/auth/identity mismatch.');
ensure(connection.baseUrl === input.baseUrl && connection.model === input.model && connection.revision === input.connectionRevision, 'Connection changed.');
const config = hub.prepare({ ...store.config, mode: 'compatible', maxSteps: 1, maxOutputTokens: 256,
contextLimit: 16000, maxRunTokens: 16000, maxTaskTokens: 16000, maxRunDurationMs: 320000,
modelResponseTimeoutMs: 300000, modelStreamIdleTimeoutMs: 120000 }, { type: 'connection', id: input.connectionId });
ensure(config.modelRouting.members.length === 1 && config.modelRouting.maxAttempts === 1 && config.maxOutputTokens === 256, 'One-request hard limits missing.');
globalThis.__messagesMProbeOnce ??= new Set();
ensure(!globalThis.__messagesMProbeOnce.has(input.sessionId), 'This sample cannot be replayed.');
globalThis.__messagesMProbeOnce.add(input.sessionId);
session.title = input.title;
const run = store.beginRun(session, input.prompt, false, config, { forceNewTask: true });
store.message(session, 'user', input.prompt);
session.providerMessages.push({ role: 'user', content: input.prompt });
const step = store.beginStep(session, run), planner = runtime.requestPlanner();
let plan, response, diagnostic = null, requestEvent, receipts = 0, wireStarts = 0;
const wireIds = new Set();
const signal = AbortSignal.timeout(320000);
let stopReason = 'identity_probe_complete';
try {
plan = planner.prepare(session, run, config, { system: 'This is a connection identity preflight. Reply to the user exactly. No tools are available. Do not perform any project work.', tools: [], stepId: step.id });
const variants = plan.body.kind === 'model-route-plan' ? plan.body.variants.map(v => v.body) : [plan.body];
ensure(variants.length === 1 && variants.every(b => !(b.tools?.length) && b.max_tokens === 256 && b.model === input.model), 'Wire request violates no-tool contract.');
ensure(plan.record.tokenReservation.reservedTokens <= 16000, 'Reservation exceeds sealed budget.');
const progress = event => {
if (event.consumption) {
receipts += Number(store.recordConsumption(session, run, { ...event.consumption, requestId: plan.record.id }, plan.record.tokenReservation.reservedTokens));
return;
}
if (event.phase === 'running' && event.wireAttemptId) { wireIds.add(event.wireAttemptId); wireStarts = wireIds.size; ensure(wireStarts === 1, 'No retries permitted.'); }
requestEvent ??= store.event(session, { kind: 'model', source: 'model_connection', label: '单次 Messages 身份预检(无工具)', phase: 'running', runId: run.id, contextPlanId: plan.record.id, startedAt: new Date().toISOString() });
if (event.modelAttemptId) requestEvent.modelAttemptId = event.modelAttemptId;
if (event.connectionId) requestEvent.connectionId = event.connectionId;
if (event.model) requestEvent.actualModel = event.model;
if (event.modelIdentity) requestEvent.modelIdentity = event.modelIdentity;
if (event.requestUrl) requestEvent.requestUrl = event.requestUrl;
if (event.diagnostic) requestEvent.diagnostic = event.diagnostic;
if (event.stream) requestEvent.stream = event.stream;
if (event.phase) requestEvent.phase = event.phase;
plan.record.requestAttempts = wireStarts; step.modelAttempts = wireStarts;
store.save(session);
};
response = await provider.complete(config, plan.messages, project, signal, progress, plan.record, [], { attempts: 1, maxBudgetAttempts: 1 });
ensure(!response.message.tool_calls?.length, 'Unsolicited tools are never executed.');
ensure(receipts === 1, 'Expected one original consumption receipt.');
planner.finish(session, plan.record, { response, attempts: wireStarts });
session.providerMessages.push(response.message); store.message(session, 'assistant', response.message.content);
const total = (run.usage?.reportedTokens ?? run.usage?.totalTokens ?? 0);
session.status = total > 16000 ? 'paused' : 'completed';
if (total > 16000) stopReason = 'task_token_limit';
} catch (cause) {
diagnostic = cause.diagnostic ?? { source: 'workbench', origin: 'local', stage: 'identity_preflight', kind: signal.aborted ? 'cancelled' : 'runtime', title: '身份预检需要核对', message: provider.redact(cause.message), occurredAt: new Date().toISOString() };
session.status = diagnostic.kind === 'model_identity' || signal.aborted ? 'paused' : 'failed';
stopReason = diagnostic.kind;
session.lastFailure = { ...diagnostic, runId: run.id };
const message = store.message(session, 'system', diagnostic.message);
session.failure = { ...session.lastFailure, messageId: message.id };
if (plan) planner.finish(session, plan.record, { error: new Error(diagnostic.message), attempts: wireStarts });
} finally {
if (requestEvent) { requestEvent.phase = diagnostic ? 'failed' : 'completed'; requestEvent.finishedAt = new Date().toISOString(); if (diagnostic) requestEvent.diagnostic = diagnostic; }
store.finishStep(session, step, diagnostic ? 'failed' : 'completed');
store.finishRun(session, run, stopReason);
}
const attempts = hub.attempts.filter(a => a.sessionId === session.id);
ensure(attempts.length <= 1 && wireStarts <= 1, 'Single-request limit exceeded.');
return { sessionId: session.id, projectId: project.id, taskId: run.taskId, runId: run.id, status: session.status,
stopReason, attempts: attempts.length, receipts, wireStarts, actualTools: run.toolCalls, usage: run.usage,
contextPlanId: plan?.record.id ?? null, diagnostic, message: response?.message ?? null };
})({"pid":41444,"port":49292,"sessionId":"a925bc92-e52d-446d-8250-8bc3b7d1d6de","projectId":"0db7cd4e-3354-4f91-851a-2143057cfd77","runtimeRoot":"D:\\Ai-project\\REA逆向\\Agent-Workbench","connectionId":"60718c65-5812-4b34-b7fb-f1e4720f17c4","credentialRevision":1,"connectionRevision":1,"baseUrl":"http://127.0.0.1:49285","model":"claude-owned-m","title":"Owned M probe match","prompt":"Reply OK owned-m. Do not use tools."})
@@ -0,0 +1,14 @@
{
"pid": 41444,
"port": 49292,
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"runtimeRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench",
"connectionId": "60718c65-5812-4b34-b7fb-f1e4720f17c4",
"credentialRevision": 1,
"connectionRevision": 1,
"baseUrl": "http://127.0.0.1:49285",
"model": "claude-owned-m",
"title": "Owned M probe match",
"prompt": "Reply OK owned-m. Do not use tools."
}
@@ -0,0 +1,38 @@
{
"sessionId": "a925bc92-e52d-446d-8250-8bc3b7d1d6de",
"projectId": "0db7cd4e-3354-4f91-851a-2143057cfd77",
"taskId": "60111d89-dc96-4bf5-a179-a72e24ee133a",
"runId": "5982a14a-9517-43d1-9b6a-d3b1fefc3b3b",
"status": "completed",
"stopReason": "identity_probe_complete",
"attempts": 1,
"receipts": 1,
"wireStarts": 1,
"actualTools": 0,
"usage": {
"requests": 1,
"inputTokens": 40,
"outputTokens": 5,
"totalTokens": 45,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"contextPlanId": "71ddf33c-340e-496a-9287-6cd71f3e9870",
"diagnostic": null,
"message": {
"role": "assistant",
"content": "OK owned-m",
"providerState": {
"protocol": "anthropic-messages",
"endpoint": "http://127.0.0.1:49285",
"model": "claude-owned-m",
"responseId": "msg_owned_m",
"content": [
{
"type": "text",
"text": "OK owned-m"
}
]
}
}
}
@@ -0,0 +1,29 @@
[
{
"route": "/v1/messages",
"body": {
"model": "claude-owned-m",
"system": [
{
"type": "text",
"text": "This is a connection identity preflight. Reply to the user exactly. No tools are available. Do not perform any project work."
}
],
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": "Reply OK owned-m. Do not use tools."
}
]
}
],
"max_tokens": 256,
"stream": true
},
"authCorrect": true,
"version": "2023-06-01"
}
]
@@ -0,0 +1,142 @@
{
"id": "26f936a4-a5d6-40f6-8e4a-23069d5150f3",
"requestId": "a97f591a-373a-41fc-a6f9-027229903ff3",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"connectionId": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"connectionRevision": 1,
"credentialRevision": 1,
"poolId": null,
"poolRevision": 1,
"model": "claude-owned-m",
"protocol": "anthropic-messages",
"streaming": true,
"endpoint": "http://127.0.0.1:59100",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"messagesAuth": "bearer",
"ordinal": 1,
"status": "failed",
"startedAt": "2026-11-03T11:06:58.214Z",
"requestBytes": 320,
"requirements": [
"text"
],
"rejections": [],
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"usage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"usageStatus": "reported",
"requestArtifactId": "ed3db6bb-b8e9-4cf0-8e71-96e4ac67c2d0",
"wireAttemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"upstreamResponse": {
"httpStatus": 200,
"requestId": "owned_m_mismatch",
"headers": {
"content-type": "text/event-stream",
"date": "Tue, 03 Nov 2026 11:06:58 GMT",
"request-id": "owned_m_mismatch"
},
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"responseUrl": "http://127.0.0.1:59100/v1/messages",
"redirected": false,
"redirectMode": "manual"
},
"partial": {
"sequence": 6,
"bytes": 727,
"textCharacters": 10,
"textPreview": "OK owned-m",
"toolFragments": 0
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"gpt-owned-other"
],
"status": "mismatch",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"scope": "sse-stream"
}
},
"stream": {
"bytes": 727,
"capturedBytes": 727,
"frames": 6,
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"terminal": true,
"terminalSeen": true,
"protocol": "anthropic-messages"
},
"responseArtifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"responseSha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"responseRedacted": false,
"diagnostic": {
"endpoint": "http://127.0.0.1:59100",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempt": 1,
"attemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"responseUrl": "http://127.0.0.1:59100/v1/messages",
"redirected": false,
"redirectMode": "manual",
"source": "model_connection",
"origin": "local",
"stage": "model_identity",
"kind": "model_identity",
"title": "模型返回型号需要核对",
"message": "请求与返回型号不一致。本次响应及用量已保留,工具未执行;核对连接或型号映射后显式续接。",
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"gpt-owned-other"
],
"status": "mismatch",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"scope": "sse-stream"
}
},
"retryable": false,
"failoverAllowed": false,
"outcome": "failed",
"occurredAt": "2026-11-03T11:06:58.846Z",
"attempts": 1,
"wireAttemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"upstreamRequestId": "owned_m_mismatch",
"retryDecision": "stream_not_replayed",
"remainingWaitBudgetMs": 299502
},
"reason": "请求与返回型号不一致。本次响应及用量已保留,工具未执行;核对连接或型号映射后显式续接。",
"httpStatus": null,
"finishedAt": "2026-11-03T11:06:59.041Z"
}
@@ -0,0 +1,753 @@
{
"schemaVersion": 1,
"executionSchemaVersion": 1,
"exportedAt": "2026-11-03T11:06:59.592Z",
"project": {
"id": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\mismatch\\project",
"name": "project",
"addedAt": "2026-11-03T11:06:57.604Z"
},
"session": {
"id": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"title": "Owned M probe mismatch",
"status": "paused",
"createdAt": "2026-11-03T11:06:57.632Z",
"updatedAt": "2026-11-03T11:06:59.326Z",
"messages": [
{
"id": "422d3b3f-4105-4be4-80c9-e1e11459f94d",
"role": "user",
"content": "Reply OK owned-m. Do not use tools.",
"createdAt": "2026-11-03T11:06:57.930Z"
},
{
"id": "c12eb7d3-1de7-4cd7-baff-d6095eb4a250",
"role": "system",
"content": "请求与返回型号不一致。本次响应及用量已保留,工具未执行;核对连接或型号映射后显式续接。",
"createdAt": "2026-11-03T11:06:59.066Z"
}
],
"providerMessages": [
{
"role": "user",
"content": "Reply OK owned-m. Do not use tools."
}
],
"events": [
{
"id": "163b0c74-2158-49c4-9038-66dbd9d1e17b",
"createdAt": "2026-11-03T11:06:58.349Z",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"kind": "model",
"source": "model_connection",
"label": "单次 Messages 身份预检(无工具)",
"phase": "failed",
"contextPlanId": "a97f591a-373a-41fc-a6f9-027229903ff3",
"startedAt": "2026-11-03T11:06:58.349Z",
"modelAttemptId": "26f936a4-a5d6-40f6-8e4a-23069d5150f3",
"connectionId": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"actualModel": "claude-owned-m",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"diagnostic": {
"endpoint": "http://127.0.0.1:59100",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempt": 1,
"attemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"responseUrl": "http://127.0.0.1:59100/v1/messages",
"redirected": false,
"redirectMode": "manual",
"source": "model_connection",
"origin": "local",
"stage": "model_identity",
"kind": "model_identity",
"title": "模型返回型号需要核对",
"message": "请求与返回型号不一致。本次响应及用量已保留,工具未执行;核对连接或型号映射后显式续接。",
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"gpt-owned-other"
],
"status": "mismatch",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"scope": "sse-stream"
}
},
"retryable": false,
"failoverAllowed": false,
"outcome": "failed",
"occurredAt": "2026-11-03T11:06:58.846Z",
"attempts": 1,
"wireAttemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"upstreamRequestId": "owned_m_mismatch",
"retryDecision": "stream_not_replayed",
"remainingWaitBudgetMs": 299502
},
"finishedAt": "2026-11-03T11:06:59.279Z"
}
],
"changes": [],
"approvals": [],
"failure": {
"endpoint": "http://127.0.0.1:59100",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempt": 1,
"attemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"responseUrl": "http://127.0.0.1:59100/v1/messages",
"redirected": false,
"redirectMode": "manual",
"source": "model_connection",
"origin": "local",
"stage": "model_identity",
"kind": "model_identity",
"title": "模型返回型号需要核对",
"message": "请求与返回型号不一致。本次响应及用量已保留,工具未执行;核对连接或型号映射后显式续接。",
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"gpt-owned-other"
],
"status": "mismatch",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"scope": "sse-stream"
}
},
"retryable": false,
"failoverAllowed": false,
"outcome": "failed",
"occurredAt": "2026-11-03T11:06:58.846Z",
"attempts": 1,
"wireAttemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"upstreamRequestId": "owned_m_mismatch",
"retryDecision": "stream_not_replayed",
"remainingWaitBudgetMs": 299502,
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"messageId": "c12eb7d3-1de7-4cd7-baff-d6095eb4a250"
},
"lastFailure": {
"endpoint": "http://127.0.0.1:59100",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempt": 1,
"attemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"responseUrl": "http://127.0.0.1:59100/v1/messages",
"redirected": false,
"redirectMode": "manual",
"source": "model_connection",
"origin": "local",
"stage": "model_identity",
"kind": "model_identity",
"title": "模型返回型号需要核对",
"message": "请求与返回型号不一致。本次响应及用量已保留,工具未执行;核对连接或型号映射后显式续接。",
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"gpt-owned-other"
],
"status": "mismatch",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"scope": "sse-stream"
}
},
"retryable": false,
"failoverAllowed": false,
"outcome": "failed",
"occurredAt": "2026-11-03T11:06:58.846Z",
"attempts": 1,
"wireAttemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"upstreamRequestId": "owned_m_mismatch",
"retryDecision": "stream_not_replayed",
"remainingWaitBudgetMs": 299502,
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0"
},
"compactions": 0,
"estimatedTokens": 115,
"usage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"archived": false,
"revision": 19,
"eventSeq": 19,
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0"
},
"execution": {
"schemaVersion": 1,
"task": {
"schemaVersion": 1,
"id": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"goal": "Reply OK owned-m. Do not use tools.",
"instructions": [],
"constraints": [],
"acceptance": [],
"status": "paused",
"previousTaskId": null,
"createdAt": "2026-11-03T11:06:57.865Z",
"updatedAt": "2026-11-03T11:06:59.326Z",
"activeRunId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"usage": {
"requests": 1,
"inputTokens": 40,
"outputTokens": 5,
"totalTokens": 45,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"budget": {
"maxTokens": 16000
},
"verification": {
"status": "not_configured",
"checkJobId": null
},
"origin": "workbench",
"contract": {
"schemaVersion": 1,
"goal": "Reply OK owned-m. Do not use tools.",
"kind": "general",
"checks": "unspecified"
},
"workspaceIdentity": {
"kind": "local-project",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\mismatch\\project"
},
"stagnation": null
},
"run": {
"schemaVersion": 1,
"id": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"engine": "workbench",
"status": "paused",
"startedAt": "2026-11-03T11:06:57.866Z",
"finishedAt": "2026-11-03T11:06:59.326Z",
"resumeOf": null,
"model": {
"mode": "compatible",
"baseUrl": "http://127.0.0.1:59100",
"model": "claude-owned-m",
"contextLimit": 16000,
"maxOutputTokens": 256,
"configRevision": 1,
"modelResponseTimeoutMs": 300000,
"modelStreamIdleTimeoutMs": 120000,
"protocol": "anthropic-messages",
"streaming": true,
"messagesAuth": "bearer",
"nativeParameters": {},
"outputTokenField": "max_tokens",
"routing": {
"type": "connection",
"id": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"revision": 1,
"strategy": "round-robin",
"sticky": true,
"maxAttempts": 1,
"queueTimeoutMs": 15000,
"members": [
{
"id": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"name": "owned M mismatch",
"brand": "自定义",
"protocol": "anthropic-messages",
"model": "claude-owned-m",
"capabilitySchema": 1,
"capabilityDeclarations": {
"text": "supported",
"tools": "supported",
"streaming": "supported",
"vision": "unsupported",
"audio": "unknown"
},
"streaming": true,
"nativeParameters": {},
"outputTokenField": "max_tokens",
"modelIdentityPolicy": "require-match",
"modelIdentityAliases": [],
"modelIdentityAliasSource": "",
"messagesAuth": "bearer",
"baseUrl": "http://127.0.0.1:59100",
"enabled": true,
"requiresKey": true,
"contextLimit": 16000,
"maxOutputTokens": 256,
"concurrency": 1,
"revision": 1,
"source": "manual",
"createdAt": "2026-11-03T11:06:57.699Z",
"updatedAt": "2026-11-03T11:06:57.699Z",
"weight": 1,
"credentialRevision": 1
}
]
}
},
"budget": {
"maxSteps": 1,
"maxDurationMs": 320000,
"maxTokens": 16000,
"maxTaskTokens": 16000
},
"usage": {
"requests": 1,
"inputTokens": 40,
"outputTokens": 5,
"totalTokens": 45,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"steps": 1,
"toolCalls": 0,
"stopReason": "model_identity",
"checkpoint": {
"stepId": "ea837cfe-5c08-4a64-9254-5a10e19b265f",
"completedToolIds": [],
"pendingToolIds": [],
"requiresVerification": []
},
"origin": "workbench",
"purpose": "execution",
"workspaceIdentity": {
"kind": "local-project",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\mismatch\\project"
},
"currentContextPlanId": "a97f591a-373a-41fc-a6f9-027229903ff3",
"consumption": [
{
"attemptId": "26f936a4-a5d6-40f6-8e4a-23069d5150f3",
"requestId": "a97f591a-373a-41fc-a6f9-027229903ff3",
"connectionId": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"connectionName": "owned M mismatch",
"model": "claude-owned-m",
"endpoint": "http://127.0.0.1:59100",
"protocol": "anthropic-messages",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"responseUrl": "http://127.0.0.1:59100/v1/messages",
"redirected": false,
"redirectMode": "manual",
"startedAt": "2026-11-03T11:06:58.348Z",
"finishedAt": "2026-11-03T11:06:58.896Z",
"outcome": "failed",
"usageStatus": "reported",
"usage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"localEstimate": {
"schemaVersion": 1,
"inputTokens": 115,
"outputTokens": 256,
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"requestBytes": 320,
"estimateMethod": "utf8-third-plus-framing-v1"
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"gpt-owned-other"
],
"status": "mismatch",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"scope": "sse-stream"
}
},
"estimatedUnknownTokens": 0
}
]
},
"workspaceIdentity": {
"kind": "local-project",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\mismatch\\project"
},
"workspaces": [],
"runs": [
{
"id": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"status": "paused",
"startedAt": "2026-11-03T11:06:57.866Z",
"finishedAt": "2026-11-03T11:06:59.326Z",
"steps": 1,
"toolCalls": 0,
"usage": {
"requests": 1,
"inputTokens": 40,
"outputTokens": 5,
"totalTokens": 45,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"stopReason": "model_identity",
"model": {
"mode": "compatible",
"baseUrl": "http://127.0.0.1:59100",
"model": "claude-owned-m",
"contextLimit": 16000,
"maxOutputTokens": 256,
"configRevision": 1,
"modelResponseTimeoutMs": 300000,
"modelStreamIdleTimeoutMs": 120000,
"protocol": "anthropic-messages",
"streaming": true,
"messagesAuth": "bearer",
"nativeParameters": {},
"outputTokenField": "max_tokens",
"routing": {
"type": "connection",
"id": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"revision": 1,
"strategy": "round-robin",
"sticky": true,
"maxAttempts": 1,
"queueTimeoutMs": 15000,
"members": [
{
"id": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"name": "owned M mismatch",
"brand": "自定义",
"protocol": "anthropic-messages",
"model": "claude-owned-m",
"capabilitySchema": 1,
"capabilityDeclarations": {
"text": "supported",
"tools": "supported",
"streaming": "supported",
"vision": "unsupported",
"audio": "unknown"
},
"streaming": true,
"nativeParameters": {},
"outputTokenField": "max_tokens",
"modelIdentityPolicy": "require-match",
"modelIdentityAliases": [],
"modelIdentityAliasSource": "",
"messagesAuth": "bearer",
"baseUrl": "http://127.0.0.1:59100",
"enabled": true,
"requiresKey": true,
"contextLimit": 16000,
"maxOutputTokens": 256,
"concurrency": 1,
"revision": 1,
"source": "manual",
"createdAt": "2026-11-03T11:06:57.699Z",
"updatedAt": "2026-11-03T11:06:57.699Z",
"weight": 1,
"credentialRevision": 1
}
]
}
},
"budget": {
"maxSteps": 1,
"maxDurationMs": 320000,
"maxTokens": 16000,
"maxTaskTokens": 16000
},
"resumeOf": null,
"purpose": "execution",
"consumption": [
{
"attemptId": "26f936a4-a5d6-40f6-8e4a-23069d5150f3",
"requestId": "a97f591a-373a-41fc-a6f9-027229903ff3",
"connectionId": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"connectionName": "owned M mismatch",
"model": "claude-owned-m",
"endpoint": "http://127.0.0.1:59100",
"protocol": "anthropic-messages",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"responseUrl": "http://127.0.0.1:59100/v1/messages",
"redirected": false,
"redirectMode": "manual",
"startedAt": "2026-11-03T11:06:58.348Z",
"finishedAt": "2026-11-03T11:06:58.896Z",
"outcome": "failed",
"usageStatus": "reported",
"usage": {
"prompt_tokens": 40,
"completion_tokens": 5,
"total_tokens": 45,
"prompt_tokens_details": {
"cached_tokens": 0,
"cache_creation_tokens": 0
},
"raw": {
"input_tokens": 40,
"output_tokens": 5
},
"protocol": "anthropic-messages"
},
"localEstimate": {
"schemaVersion": 1,
"inputTokens": 115,
"outputTokens": 256,
"requestSha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"requestBytes": 320,
"estimateMethod": "utf8-third-plus-framing-v1"
},
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"gpt-owned-other"
],
"status": "mismatch",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"scope": "sse-stream"
}
},
"estimatedUnknownTokens": 0
}
],
"workspaceIdentity": {
"kind": "local-project",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"root": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac\\mismatch\\project"
}
}
],
"progress": {
"totalTools": 0,
"completedTools": 0,
"requiresVerification": []
},
"artifacts": [
{
"id": "509b3d12-29d4-4e17-96a9-7f86d711c211",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"kind": "model-request",
"mediaType": "application/json",
"sha256": "c495698e007ae3a30c44bdfc2f5124336c739585e52fe9c655df644073766194",
"bytes": 699,
"relative": "c4/c495698e007ae3a30c44bdfc2f5124336c739585e52fe9c655df644073766194.blob",
"createdAt": "2026-11-03T11:06:58.089Z",
"status": "stored"
},
{
"id": "c63b96d3-e062-4716-8881-1bc4c20f102a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"kind": "context-plan",
"mediaType": "application/json",
"sha256": "95e61745f3908ce19116c944f3d123a4d0f6a6c6cf901719b9ad2b5c4a8a234a",
"bytes": 2083,
"relative": "95/95e61745f3908ce19116c944f3d123a4d0f6a6c6cf901719b9ad2b5c4a8a234a.blob",
"createdAt": "2026-11-03T11:06:58.130Z",
"status": "stored"
},
{
"id": "ed3db6bb-b8e9-4cf0-8e71-96e4ac67c2d0",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"kind": "model-attempt-request",
"mediaType": "application/json",
"sha256": "341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85",
"bytes": 320,
"relative": "34/341fe2bb18b1d1fb9a6fdeadaac9d1b1568d4f134d7e74e7b22f3adf544edc85.blob",
"createdAt": "2026-11-03T11:06:58.256Z",
"status": "stored"
},
{
"id": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"kind": "model-stream-response",
"mediaType": "text/event-stream",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"bytes": 727,
"relative": "41/419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a.blob",
"createdAt": "2026-11-03T11:06:58.796Z",
"status": "stored"
}
],
"checks": [],
"browsingTabs": [],
"browsingActions": [],
"previewPlans": [],
"previewJobs": [],
"mcpJobs": [],
"taskPlans": [],
"taskProofs": [],
"researchTargets": [],
"researchJobs": [],
"engineJobs": [],
"browserPlans": [],
"browserJobs": [],
"contextPlans": [
{
"schemaVersion": 1,
"id": "a97f591a-373a-41fc-a6f9-027229903ff3",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"stepId": "ea837cfe-5c08-4a64-9254-5a10e19b265f",
"requestIndex": 1,
"status": "failed",
"createdAt": "2026-11-03T11:06:58.089Z",
"finishedAt": "2026-11-03T11:06:59.224Z",
"requestSha256": "c495698e007ae3a30c44bdfc2f5124336c739585e52fe9c655df644073766194",
"requestBytes": 699,
"estimatedInputTokens": 115,
"requestArtifactId": "509b3d12-29d4-4e17-96a9-7f86d711c211",
"tokenReservation": {
"schemaVersion": 1,
"method": "observed-route-max-v1",
"overheadTokens": 0,
"reservedTokens": 371,
"routes": [
{
"connectionId": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"model": "claude-owned-m",
"protocol": "anthropic-messages",
"endpoint": "http://127.0.0.1:59100",
"localInputTokens": 115,
"outputReservation": 256,
"inputRatio": 1,
"observedOutputTokens": 256,
"observedTotalFloor": 0,
"reservedTokens": 371,
"sources": []
}
]
},
"maxBudgetAttempts": 8,
"estimateMethod": "maximum-wire-variant-utf8-v1",
"beforeInputTokens": 115,
"beforeBytes": 699,
"budget": {
"contextLimit": 16000,
"outputReservation": 256,
"safetyTokens": 1600,
"inputBudget": 14144,
"targetInputTokens": 14144,
"runRemainingTokens": 16000,
"taskRemainingTokens": 16000,
"limitingReason": "context_limit"
},
"mandatoryInputTokens": 115,
"nextReadFixedInputTokens": 115,
"workingRetentionTokens": 0,
"workingRetentionPolicyVersion": 1,
"originalMessages": 1,
"retainedMessages": 1,
"archivedGroups": 0,
"retainedGroups": [
0
],
"historyIndexArtifactId": null,
"historyIndexSha256": null,
"actualUsage": null,
"workingSetArtifactId": null,
"workingSetSha256": null,
"workingSetEntries": 0,
"workingSetViewVersion": 2,
"latestToolsArtifactId": null,
"latestToolsSha256": null,
"latestToolsEntries": 0,
"projectMemoryArtifactId": null,
"projectMemorySha256": null,
"projectMemoryEntries": 0,
"skillCatalogArtifactId": null,
"skillCatalogSha256": null,
"skillCatalogEntries": 0,
"mcpCatalogArtifactId": null,
"mcpCatalogSha256": null,
"mcpCatalogEntries": 0,
"requestAttempts": 1,
"responseSha256": null,
"source": "workbench",
"artifactId": "c63b96d3-e062-4716-8881-1bc4c20f102a",
"error": "请求与返回型号不一致。本次响应及用量已保留,工具未执行;核对连接或型号映射后显式续接。"
}
],
"changeSets": [],
"plans": [],
"deliveries": [],
"deliveryPlans": [],
"accountedTaskTokens": 45
}
}
@@ -0,0 +1,83 @@
(async function nativeMessagesProbe(input) {
const ensure = (value, message) => { if (!value) throw new Error(message); };
ensure(process.pid === input.pid, 'Unexpected host PID.');
const servers = process._getActiveHandles().filter(h => h?.listening && h.address?.()?.address === '127.0.0.1' && h.address?.()?.port === input.port);
ensure(servers.length === 1, 'Expected one owned loopback server.');
const context = servers[0].workbenchHostAuthority?.context;
ensure(context?.provider && context?.store && context?.runtime, 'Original context missing.');
const { store, provider, runtime } = context;
ensure(runtime.appDirectory === input.runtimeRoot, 'Loaded runtime changed.');
ensure(![...store.sessions.values()].some(s => ['running', 'stopping', 'awaiting_approval'].includes(s.status)), 'Host is busy.');
const session = store.session(input.sessionId), project = store.project(session.projectId);
ensure(session.projectId === input.projectId && session.messages.length === 0 && !session.runId, 'Probe session must be new and owned.');
const hub = provider.modelHub(store);
const connection = hub.view().connections.find(c => c.id === input.connectionId);
ensure(connection?.hasKey && connection.credentialRevision === input.credentialRevision, 'Original saved credential revision changed.');
ensure(connection.protocol === 'anthropic-messages' && connection.messagesAuth === 'bearer' && connection.modelIdentityPolicy === 'require-match', 'Probe protocol/auth/identity mismatch.');
ensure(connection.baseUrl === input.baseUrl && connection.model === input.model && connection.revision === input.connectionRevision, 'Connection changed.');
const config = hub.prepare({ ...store.config, mode: 'compatible', maxSteps: 1, maxOutputTokens: 256,
contextLimit: 16000, maxRunTokens: 16000, maxTaskTokens: 16000, maxRunDurationMs: 320000,
modelResponseTimeoutMs: 300000, modelStreamIdleTimeoutMs: 120000 }, { type: 'connection', id: input.connectionId });
ensure(config.modelRouting.members.length === 1 && config.modelRouting.maxAttempts === 1 && config.maxOutputTokens === 256, 'One-request hard limits missing.');
globalThis.__messagesMProbeOnce ??= new Set();
ensure(!globalThis.__messagesMProbeOnce.has(input.sessionId), 'This sample cannot be replayed.');
globalThis.__messagesMProbeOnce.add(input.sessionId);
session.title = input.title;
const run = store.beginRun(session, input.prompt, false, config, { forceNewTask: true });
store.message(session, 'user', input.prompt);
session.providerMessages.push({ role: 'user', content: input.prompt });
const step = store.beginStep(session, run), planner = runtime.requestPlanner();
let plan, response, diagnostic = null, requestEvent, receipts = 0, wireStarts = 0;
const wireIds = new Set();
const signal = AbortSignal.timeout(320000);
let stopReason = 'identity_probe_complete';
try {
plan = planner.prepare(session, run, config, { system: 'This is a connection identity preflight. Reply to the user exactly. No tools are available. Do not perform any project work.', tools: [], stepId: step.id });
const variants = plan.body.kind === 'model-route-plan' ? plan.body.variants.map(v => v.body) : [plan.body];
ensure(variants.length === 1 && variants.every(b => !(b.tools?.length) && b.max_tokens === 256 && b.model === input.model), 'Wire request violates no-tool contract.');
ensure(plan.record.tokenReservation.reservedTokens <= 16000, 'Reservation exceeds sealed budget.');
const progress = event => {
if (event.consumption) {
receipts += Number(store.recordConsumption(session, run, { ...event.consumption, requestId: plan.record.id }, plan.record.tokenReservation.reservedTokens));
return;
}
if (event.phase === 'running' && event.wireAttemptId) { wireIds.add(event.wireAttemptId); wireStarts = wireIds.size; ensure(wireStarts === 1, 'No retries permitted.'); }
requestEvent ??= store.event(session, { kind: 'model', source: 'model_connection', label: '单次 Messages 身份预检(无工具)', phase: 'running', runId: run.id, contextPlanId: plan.record.id, startedAt: new Date().toISOString() });
if (event.modelAttemptId) requestEvent.modelAttemptId = event.modelAttemptId;
if (event.connectionId) requestEvent.connectionId = event.connectionId;
if (event.model) requestEvent.actualModel = event.model;
if (event.modelIdentity) requestEvent.modelIdentity = event.modelIdentity;
if (event.requestUrl) requestEvent.requestUrl = event.requestUrl;
if (event.diagnostic) requestEvent.diagnostic = event.diagnostic;
if (event.stream) requestEvent.stream = event.stream;
if (event.phase) requestEvent.phase = event.phase;
plan.record.requestAttempts = wireStarts; step.modelAttempts = wireStarts;
store.save(session);
};
response = await provider.complete(config, plan.messages, project, signal, progress, plan.record, [], { attempts: 1, maxBudgetAttempts: 1 });
ensure(!response.message.tool_calls?.length, 'Unsolicited tools are never executed.');
ensure(receipts === 1, 'Expected one original consumption receipt.');
planner.finish(session, plan.record, { response, attempts: wireStarts });
session.providerMessages.push(response.message); store.message(session, 'assistant', response.message.content);
const total = (run.usage?.reportedTokens ?? run.usage?.totalTokens ?? 0);
session.status = total > 16000 ? 'paused' : 'completed';
if (total > 16000) stopReason = 'task_token_limit';
} catch (cause) {
diagnostic = cause.diagnostic ?? { source: 'workbench', origin: 'local', stage: 'identity_preflight', kind: signal.aborted ? 'cancelled' : 'runtime', title: '身份预检需要核对', message: provider.redact(cause.message), occurredAt: new Date().toISOString() };
session.status = diagnostic.kind === 'model_identity' || signal.aborted ? 'paused' : 'failed';
stopReason = diagnostic.kind;
session.lastFailure = { ...diagnostic, runId: run.id };
const message = store.message(session, 'system', diagnostic.message);
session.failure = { ...session.lastFailure, messageId: message.id };
if (plan) planner.finish(session, plan.record, { error: new Error(diagnostic.message), attempts: wireStarts });
} finally {
if (requestEvent) { requestEvent.phase = diagnostic ? 'failed' : 'completed'; requestEvent.finishedAt = new Date().toISOString(); if (diagnostic) requestEvent.diagnostic = diagnostic; }
store.finishStep(session, step, diagnostic ? 'failed' : 'completed');
store.finishRun(session, run, stopReason);
}
const attempts = hub.attempts.filter(a => a.sessionId === session.id);
ensure(attempts.length <= 1 && wireStarts <= 1, 'Single-request limit exceeded.');
return { sessionId: session.id, projectId: project.id, taskId: run.taskId, runId: run.id, status: session.status,
stopReason, attempts: attempts.length, receipts, wireStarts, actualTools: run.toolCalls, usage: run.usage,
contextPlanId: plan?.record.id ?? null, diagnostic, message: response?.message ?? null };
})({"pid":56872,"port":59101,"sessionId":"fd617a07-ec7a-43fb-9316-708b8c0e00f2","projectId":"bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997","runtimeRoot":"D:\\Ai-project\\REA逆向\\Agent-Workbench","connectionId":"cac6f2de-c65f-4411-83ab-6714edb03fc6","credentialRevision":1,"connectionRevision":1,"baseUrl":"http://127.0.0.1:59100","model":"claude-owned-m","title":"Owned M probe mismatch","prompt":"Reply OK owned-m. Do not use tools."})
@@ -0,0 +1,14 @@
{
"pid": 56872,
"port": 59101,
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"runtimeRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench",
"connectionId": "cac6f2de-c65f-4411-83ab-6714edb03fc6",
"credentialRevision": 1,
"connectionRevision": 1,
"baseUrl": "http://127.0.0.1:59100",
"model": "claude-owned-m",
"title": "Owned M probe mismatch",
"prompt": "Reply OK owned-m. Do not use tools."
}
@@ -0,0 +1,66 @@
{
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"projectId": "bcc81b15-7c1f-4ff3-beb0-43b6fd2fc997",
"taskId": "663b0880-a11e-4da8-a7b2-a8b25ecc18a2",
"runId": "4796ee46-eabf-4c0f-88ff-d55d11c854e0",
"status": "paused",
"stopReason": "model_identity",
"attempts": 1,
"receipts": 1,
"wireStarts": 1,
"actualTools": 0,
"usage": {
"requests": 1,
"inputTokens": 40,
"outputTokens": 5,
"totalTokens": 45,
"unknownRequests": 0,
"estimatedUnknownTokens": 0
},
"contextPlanId": "a97f591a-373a-41fc-a6f9-027229903ff3",
"diagnostic": {
"endpoint": "http://127.0.0.1:59100",
"requestUrl": "http://127.0.0.1:59100/v1/messages",
"model": "claude-owned-m",
"route": "/v1/messages",
"attempt": 1,
"attemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"responseUrl": "http://127.0.0.1:59100/v1/messages",
"redirected": false,
"redirectMode": "manual",
"source": "model_connection",
"origin": "local",
"stage": "model_identity",
"kind": "model_identity",
"title": "模型返回型号需要核对",
"message": "请求与返回型号不一致。本次响应及用量已保留,工具未执行;核对连接或型号映射后显式续接。",
"modelIdentity": {
"schemaVersion": 1,
"requestedModel": "claude-owned-m",
"reportedModels": [
"gpt-owned-other"
],
"status": "mismatch",
"complete": true,
"invalidIdentifiers": false,
"policy": "require-match",
"scope": "Identifiers in the outgoing request and upstream response; neither equal identifiers nor a declared mapping authenticate model weights.",
"responseSource": {
"artifactId": "236f5327-1744-4aa4-84b1-de9931aa082e",
"sha256": "419bbf33a9232a8de864bb016bf79f4ba1152558f46cfe2fbca33974b456197a",
"sessionId": "fd617a07-ec7a-43fb-9316-708b8c0e00f2",
"scope": "sse-stream"
}
},
"retryable": false,
"failoverAllowed": false,
"outcome": "failed",
"occurredAt": "2026-11-03T11:06:58.846Z",
"attempts": 1,
"wireAttemptId": "54da95df-03b2-439b-a5b6-f7c271bb2e59",
"upstreamRequestId": "owned_m_mismatch",
"retryDecision": "stream_not_replayed",
"remainingWaitBudgetMs": 299502
},
"message": null
}
@@ -0,0 +1,29 @@
[
{
"route": "/v1/messages",
"body": {
"model": "claude-owned-m",
"system": [
{
"type": "text",
"text": "This is a connection identity preflight. Reply to the user exactly. No tools are available. Do not perform any project work."
}
],
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": "Reply OK owned-m. Do not use tools."
}
]
}
],
"max_tokens": 256,
"stream": true
},
"authCorrect": true,
"version": "2023-06-01"
}
]
@@ -0,0 +1,30 @@
{
"passed": true,
"realModelRequests": 0,
"results": [
{
"mode": "match",
"passed": true,
"requests": 1,
"actualTools": 0,
"status": "completed",
"inspectorClosed": true
},
{
"mode": "mismatch",
"passed": true,
"requests": 1,
"actualTools": 0,
"status": "paused",
"inspectorClosed": true
},
{
"mode": "429",
"passed": true,
"requests": 1,
"actualTools": 0,
"status": "failed",
"inspectorClosed": true
}
]
}
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,26 @@
{
"success": true,
"results": [
{
"script": "verify-model-center-ui",
"code": 0,
"log": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\model-selection-k-ui-regression-86916c34-8568-4e02-beaf-26c5400969e0\\verify-model-center-ui.log"
},
{
"script": "verify-model-protocols-ui",
"code": 0,
"log": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\model-selection-k-ui-regression-86916c34-8568-4e02-beaf-26c5400969e0\\verify-model-protocols-ui.log"
},
{
"script": "verify-model-identity-ui",
"code": 0,
"log": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\model-selection-k-ui-regression-86916c34-8568-4e02-beaf-26c5400969e0\\verify-model-identity-ui.log"
},
{
"script": "verify-model-selection-ui",
"code": 0,
"log": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\model-selection-k-ui-regression-86916c34-8568-4e02-beaf-26c5400969e0\\verify-model-selection-ui.log"
}
],
"realModelRequests": 0
}
@@ -0,0 +1,7 @@
PASS connection form stores only the volatile credential and renders configured status without exposing it
PASS pool editor saves weighted members and binds selection for new Runs
PASS schema-bound import preview, opt-in automatic sync, revision update and sync shutdown keep credentials hidden
PASS real historical pagination detects the owned corrupted older artifact, retains its identity and leaves original bytes unchanged
PASS local inspection shows real database and artifact checks and emits zero model requests
PASS 390px model-center modal and connection cards remain operable without horizontal page overflow
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\model-center-ui-28d10c07-ff2e-4d62-a2cf-adfcf3d06215
@@ -0,0 +1,4 @@
PASS strict policy and multiple declared aliases save and reload at desktop and 390px without a model call
PASS strict mismatch pauses before any proposal, retains reported usage, persists on refresh and downloads SHA-verified original
PASS physical request ledger retains the same identity and source without page errors
{"success":true,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\model-identity-ui-60f7e748-595b-4885-b007-137afbfef9cc","checks":3}
@@ -0,0 +1,6 @@
PASS actual form saves Responses and declared stream, refresh preserves both and narrow editor remains operable
PASS separate text, nonce tool and actual SSE probes bind only their requested capability and own attempts
PASS actual pooled task excludes text-only before dispatch, executes original tool once and downloads SHA-matched original SSE
PASS desktop and actual paired gateway show same bounded stream; mobile stops original HTTP with zero fragment tool dispatch or replay
PASS truncated output keeps explicit unknown ledger, partial evidence and no automatic failover or incomplete tool execution
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\model-protocols-ui-def9c2b5-3a45-45af-b027-f89493263986
@@ -0,0 +1,8 @@
PASS sidebar and header share the authoritative multi-model selector; selection persists without a request
PASS add from primary entry keeps the previous default, supports a new selection and survives reload without paid probes
PASS budget-only saves preserve the pool and original connection editing is explicitly secondary and cancellable
PASS actual pool request reaches only the selected endpoint; changing defaults preserves the sealed old Run and shows both identities
PASS opening refreshes another tab selection and choosing a real connection exits demo without losing profiles
PASS 390px header and sidebar both add and switch with long model labels and no page overflow
PASS active request locks route and add controls and server rejects budget overrides across sessions
{"success":true,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\model-selection-k-83993012-a896-432c-8026-cb05dc7fc318","checks":7}
Binary file not shown.

After

Width:  |  Height:  |  Size: 138 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 141 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 40 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 50 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 58 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 55 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 53 KiB

File diff suppressed because one or more lines are too long
@@ -0,0 +1 @@
{"success": true, "version": "1.24.4", "testedAt": "2026-11-03T11:26:05.017312+00:00", "archives": [{"kind": "portable", "path": "D:\\Ai-project\\REA\u9006\u5411\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103112054864.zip", "files": 10936, "crcPassed": true, "manifestAndDirectoryMatch": true, "sha256": "31983cb4258ebcf517ee66f5da2b4e0464cc0e04620341a2491b38b5d56ce329"}, {"kind": "source", "path": "D:\\Ai-project\\REA\u9006\u5411\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-source-20261103112054864.zip", "files": 30124, "crcPassed": true, "manifestAndDirectoryMatch": true, "sha256": "20c6c851480abaec78de0860885f41370e27773863ae96d0dac97f0e437d51d0"}, {"kind": "evidence", "path": "D:\\Ai-project\\REA\u9006\u5411\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-evidence-20261103112054864.zip", "files": 18877, "crcPassed": true, "manifestAndDirectoryMatch": true, "sha256": "3b3ea76e98739b657fda91bdb1d95e60f451346f9d6453125a53aa6dc7ae4ef2"}], "extractionRoot": null}
@@ -0,0 +1 @@
{"success": true, "version": "1.24.4", "testedAt": "2026-11-03T11:04:38.682153+00:00", "archives": [{"kind": "portable", "path": "D:\\Ai-project\\REA\u9006\u5411\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103105910315.zip", "files": 10936, "crcPassed": true, "manifestAndDirectoryMatch": true, "sha256": "78256102444dd5030f8767e66d54d776a8a788c6eca02d804815fa24c2db2c53"}, {"kind": "source", "path": "D:\\Ai-project\\REA\u9006\u5411\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-source-20261103105910315.zip", "files": 30124, "crcPassed": true, "manifestAndDirectoryMatch": true, "sha256": "22854813be9920547d88ca71313af73889a654e817fd3613747f588c33429f9c"}, {"kind": "evidence", "path": "D:\\Ai-project\\REA\u9006\u5411\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-evidence-20261103105910315.zip", "files": 18877, "crcPassed": true, "manifestAndDirectoryMatch": true, "sha256": "26ab3e926b39d70975b528980142eaba2fec10b941b1ef6484b5be08d656ef2e"}], "extractionRoot": null}
@@ -0,0 +1,12 @@
{
"stage": "before-formal-upgrade",
"readOnly": true,
"inventory": "IAB 2; existing Workbench tab 8 at release=1.23.0",
"error": "browser.user is unavailable.",
"attempts": ["cua.getTab existing tab", "same browser documented agent.browsers.get and tabs.get"],
"guidanceRead": "browser-troubleshooting",
"actionsOnUserPage": 0,
"recovered": true,
"recovery": "One CUA reset; same IAB 2 and existing tab 8 recovered. Observed formal v1.24.3 and original J strict pause; no input draft and no page mutation.",
"next": "After upgrade, use this recovered tab for formal v1.24.4 review. Native Sky and independent compiled UI remain separate checks."
}
@@ -0,0 +1,20 @@
> local-agent-workbench@1.24.4 build
> tsc --noEmit && vite build && node scripts/build-mobile.mjs && node scripts/build-server.mjs
vite v7.3.7 building client environment for production...
transforming...
✓ 1759 modules transformed.
rendering chunks...
computing gzip size...
dist-client/index.html 0.60 kB │ gzip: 0.41 kB
dist-client/.vite/manifest.json 0.65 kB │ gzip: 0.24 kB
dist-client/mobile.html 0.84 kB │ gzip: 0.47 kB
dist-client/assets/mobile-joLHctQw.css 12.39 kB │ gzip: 3.22 kB
dist-client/assets/desktop-D67Tn6c_.css 168.00 kB │ gzip: 31.73 kB
dist-client/assets/mobile-4zqcwTeD.js 27.19 kB │ gzip: 9.34 kB
dist-client/assets/approval-presentation-Chut4EFA.js 226.80 kB │ gzip: 71.65 kB
dist-client/assets/desktop-D2oO9iie.js 404.17 kB │ gzip: 114.60 kB
✓ built in 1.20s
Mobile shell built: 3 allowed assets; API and project data are never cached.
Server bundle built.
@@ -0,0 +1,20 @@
> local-agent-workbench@1.24.4 build
> tsc --noEmit && vite build && node scripts/build-mobile.mjs && node scripts/build-server.mjs
vite v7.3.7 building client environment for production...
transforming...
✓ 1759 modules transformed.
rendering chunks...
computing gzip size...
dist-client/index.html 0.60 kB │ gzip: 0.41 kB
dist-client/.vite/manifest.json 0.65 kB │ gzip: 0.24 kB
dist-client/mobile.html 0.84 kB │ gzip: 0.47 kB
dist-client/assets/mobile-joLHctQw.css 12.39 kB │ gzip: 3.22 kB
dist-client/assets/desktop-D67Tn6c_.css 168.00 kB │ gzip: 31.73 kB
dist-client/assets/mobile-4zqcwTeD.js 27.19 kB │ gzip: 9.34 kB
dist-client/assets/approval-presentation-Chut4EFA.js 226.80 kB │ gzip: 71.65 kB
dist-client/assets/desktop-BVDBowt7.js 404.50 kB │ gzip: 114.67 kB
✓ built in 1.24s
Mobile shell built: 3 allowed assets; API and project data are never cached.
Server bundle built.
@@ -0,0 +1,20 @@
> local-agent-workbench@1.24.3 build
> tsc --noEmit && vite build && node scripts/build-mobile.mjs && node scripts/build-server.mjs
vite v7.3.7 building client environment for production...
transforming...
✓ 1759 modules transformed.
rendering chunks...
computing gzip size...
dist-client/index.html 0.60 kB │ gzip: 0.41 kB
dist-client/.vite/manifest.json 0.65 kB │ gzip: 0.24 kB
dist-client/mobile.html 0.84 kB │ gzip: 0.47 kB
dist-client/assets/mobile-joLHctQw.css 12.39 kB │ gzip: 3.22 kB
dist-client/assets/desktop-D67Tn6c_.css 168.00 kB │ gzip: 31.73 kB
dist-client/assets/mobile-4zqcwTeD.js 27.19 kB │ gzip: 9.34 kB
dist-client/assets/approval-presentation-Chut4EFA.js 226.80 kB │ gzip: 71.65 kB
dist-client/assets/desktop-D2oO9iie.js 404.17 kB │ gzip: 114.60 kB
✓ built in 1.21s
Mobile shell built: 3 allowed assets; API and project data are never cached.
Server bundle built.
@@ -0,0 +1,20 @@
> local-agent-workbench@1.24.4 build
> tsc --noEmit && vite build && node scripts/build-mobile.mjs && node scripts/build-server.mjs
vite v7.3.7 building client environment for production...
transforming...
✓ 1759 modules transformed.
rendering chunks...
computing gzip size...
dist-client/index.html 0.60 kB │ gzip: 0.41 kB
dist-client/.vite/manifest.json 0.65 kB │ gzip: 0.24 kB
dist-client/mobile.html 0.84 kB │ gzip: 0.47 kB
dist-client/assets/mobile-joLHctQw.css 12.39 kB │ gzip: 3.22 kB
dist-client/assets/desktop-D67Tn6c_.css 168.00 kB │ gzip: 31.73 kB
dist-client/assets/mobile-4zqcwTeD.js 27.19 kB │ gzip: 9.34 kB
dist-client/assets/approval-presentation-Chut4EFA.js 226.80 kB │ gzip: 71.65 kB
dist-client/assets/desktop-BVDBowt7.js 404.50 kB │ gzip: 114.67 kB
✓ built in 1.21s
Mobile shell built: 3 allowed assets; API and project data are never cached.
Server bundle built.
@@ -0,0 +1,156 @@
✔ Messages builds native system/tool history, bounds images, and seals a distinct mixed-pool variant (6.7516ms)
✔ Messages URL and native parameter contracts preserve other protocol paths and hard output budget (0.524ms)
✔ native usage includes both cache buckets once, retains raw and rejects invalid counters (0.1777ms)
✔ real HTTP Messages JSON executes one tool, retains opaque history, actual URL and cache accounting (3013.2609ms)
✔ real HTTP Messages SSE executes one tool, retains opaque history, actual URL and cache accounting (2190.1543ms)
✔ Messages api-key authentication is explicit; mismatched model remains blocked with original usage (5.1733ms)
✔ Messages bearer authentication is explicit; mismatched model remains blocked with original usage (4.0284ms)
✔ Messages redirect is not followed, either credential stays at the original server (3.9361ms)
✔ strict native SSE reads the nested start model and retains mismatch source without retries (3.4398ms)
✔ a closed native tool block cannot execute before message_stop, then executes exactly once (2543.4174ms)
✔ mixed Chat/Messages pool seals both variants and preserves one completed tool on failover (3149.3977ms)
✔ native JSON rejects incomplete stops, duplicate tools and hosted content (0.3491ms)
▶ strict SSE validates all blocks, terminal, event names, UTF8 and tool JSON before dispatch
✔ reject native stream 0 (1.1702ms)
✔ reject native stream 1 (0.2582ms)
✔ reject native stream 2 (0.2249ms)
✔ reject native stream 3 (0.2055ms)
✔ reject native stream 4 (0.1722ms)
✔ reject native stream 5 (0.2107ms)
✔ reject native stream 6 (0.1566ms)
✔ reject native stream 7 (0.1862ms)
✔ reject native stream 8 (0.3112ms)
✔ strict SSE validates all blocks, terminal, event names, UTF8 and tool JSON before dispatch (3.6467ms)
✔ upstream SSE error retains original type/request ID and initial usage, never retries (6.7765ms)
✔ stop cancels the owned stream and retains partial consumption without running tool or replaying (60.2562ms)
✔ saved Messages import preserves auth choice, rejects endpoint duplication, and keeps source file untouched (533.9392ms)
✔ a Windows virtual data-directory name binds new storage and credentials to its actual physical root (2038.9747ms)
✔ physical data-root adoption still rejects explicit junctions at the directory or any ancestor (668.77ms)
✔ actual Windows CurrentUser DPAPI roundtrips owned UTF-8 and rejects damaged bytes and wrong entropy (1500.2543ms)
✔ only explicitly supplied keys persist; new provider restores original revisions and public views contain no key (2398.5932ms)
✔ registry and cipher commit failure restores original routing, keys, block and every database row (1696.1901ms)
✔ a two-key import with a protection failure rolls back all registry and cipher entries and retains its preview (1964.8414ms)
✔ connection revision changes retain the original saved block and report a source mismatch on restart (2246.5277ms)
✔ disable skips real restart loading, explicit enable restores, delete preserves memory and forget removes both (2828.5029ms)
✔ corrupt saved block and another SID do not load; failures are visible without returning ciphertext or plaintext (2543.9984ms)
✔ legacy persistence requires a new explicit key and binds config and credential revisions; stale requests cannot save (2579.6184ms)
✔ owned plaintext keys never reach SQLite pages, WAL, config or public execution exports (1401.7126ms)
✔ schema1 connections remain readable without inventing observed capabilities or migrating original state (562.0547ms)
✔ protocol, unsupported capabilities and stream choice persist; client claims cannot forge verified evidence (955.8479ms)
✔ ability evidence belongs to exact connection and credential revision; unknown and unsupported reject required routing (0.2814ms)
✔ invalid declarations and inconsistent stream settings reject before registry, revision and credentials mutate (500.5654ms)
✔ model registry keeps original credentials/endpoint intact, rejects stale edits, and forgets only volatile keys on restart (838.9311ms)
✔ config import previews exact revisions, keeps secrets volatile, syncs only when enabled and preserves manual edits on conflicts (912.9089ms)
✔ real HTTP pool fails over 429 to another model and never repeats completed tools; attempts seal actual request bytes and Run identity (2941.1691ms)
✔ pool rotation and weights select members on new Runs; disabled and missing credentials are excluded without requests (3703.9223ms)
✔ 401 blocks the credential revision; updating the key restores availability without leaking error echoes (1884.1572ms)
✔ queued request cancellation and in-flight disable release only owned leases and do not dispatch another HTTP request (1058.8129ms)
✔ compiled model-center HTTP API persists public profiles, blocks forged origins, and local operations never invoke a model (2194.2978ms)
✔ model registry disk failures restore routing, pool profiles and credential references without partial updates (778.8009ms)
✔ model selection and saved configuration commit together and failures preserve the original mode and endpoint (924.2998ms)
✔ configuration imports roll back the entire batch and retain one usable preview after validation or commit failure (729.3175ms)
✔ HTTP quota retains native code/type/param/request id, bounded original and allowlisted headers, never repeats that connection (2046.2917ms)
✔ Retry-After above 90 seconds remains exact and a short local budget cannot trigger an early retry (1.8063ms)
✔ Retry-After date, expired date, huge value and invalid header retain explicit semantics (1.1943ms)
✔ unclassified 429 remains unspecified and every bounded actual retry retains its own original error (2049.0124ms)
✔ oversized HTTP error saves exactly the bounded prefix and labels its incomplete range (2072.3122ms)
✔ Chat SSE error preserves upstream fields, original failed event, partial tool and reported usage without executing or retrying (2192.2207ms)
✔ Responses failed SSE retains original event/response identity and usage (1856.3933ms)
✔ Responses incomplete SSE retains original event/response identity and usage (1965.2924ms)
✔ unsupported successful Responses output retains original JSON and usage with a local protocol origin (2218.8801ms)
✔ HTTP 200 native error is still an upstream failure rather than a fabricated successful answer (1.4139ms)
✔ JSON-escaped credential echoes are redacted in the original preview and capture as well as parsed fields (0.917ms)
✔ quota failover uses another independent pool member and never replays a failed stream (2913.9022ms)
✔ native parameters are isolated by pool member, preserve Responses state and bind the original output reservation (0.6217ms)
✔ unsupported native fields and invalid values reject before dispatch instead of silently dropping or overriding constraints (0.795ms)
✔ native settings for legacy and registered profiles persist without changing credentials or old records and reject stale edits (1044.1796ms)
✔ compiled native settings API saves/restores exact parameters and rejects reserved fields without mutating the profile (2529.4652ms)
✔ legacy connection test sends its saved native fields and selected output limit field without changing the configuration (2151.2787ms)
✔ identity distinguishes missing, invalid, mixed, exact and explicitly declared aliases without guessing (1.3064ms)
✔ chat-completions JSON rejects a different model before tool dispatch, retaining source and reported usage (1988.3643ms)
✔ chat-completions SSE rejects a different model before tool dispatch, retaining source and reported usage (1999.8304ms)
✔ responses JSON rejects a different model before tool dispatch, retaining source and reported usage (2363.2123ms)
✔ responses SSE rejects a different model before tool dispatch, retaining source and reported usage (2234.5137ms)
✔ record policy accepts missing identity but keeps it explicit, and strict declared mapping is frozen in the consumption (2.6047ms)
✔ mixed streamed identities cannot pass strict mode, and interrupted streams retain observed identifiers without authenticating them (1.4246ms)
✔ identity refusal never fails over to another pool member and raw response model cannot leak known credentials (2214.1019ms)
✔ strict refusal remains visible even when the same original response exhausts the budget (1284.4646ms)
✔ strict identity crosses the controlled engine bridge with its original source and zero tool dispatch (12058.5436ms)
✔ frozen direct connection policy and delivery reports preserve response identity without filling historical gaps (2538.228ms)
✔ saved engine budgets verify all Chat/Responses route variants and reject a mismatched output allowance (718.3841ms)
✔ owned CC Switch SQLite imports the selected TOML Responses wire_api and leaves unknown clients untouched (895.1361ms)
✔ Responses executes one original tool through Runtime, carries encrypted Items, reports raw usage and seals exact wire bytes (2086.0016ms)
✔ mixed-protocol pool keeps both actual variants and changes route on 429 without re-executing completed tools (2687.1053ms)
▶ Responses incomplete, failed, missing call identity and unsupported hosted outputs never dispatch a tool
✔ Reject {"body":{"id":"resp_owned","status":"incomplete","output":[{"type":"function_call","id":"fc_partial" (1713.742ms)
✔ Reject {"body":{"id":"resp_owned","status":"failed","output":[{"type":"function_call","id":"fc_failed","cal (1739.4353ms)
✔ Reject {"body":{"id":"resp_owned","status":"completed","output":[{"type":"function_call","id":"fc_bad","cal (1618.842ms)
✔ Reject {"body":{"id":"resp_owned","status":"completed","output":[{"type":"computer_call","id":"not-supporte (1903.8158ms)
✔ Responses incomplete, failed, missing call identity and unsupported hosted outputs never dispatch a tool (6976.7273ms)
▶ invalid JSON in a nonstream Chat or Responses tool group prevents every tool and pool replay
✔ chat-completions (2636.4581ms)
✔ responses (2974.6155ms)
✔ invalid JSON in a nonstream Chat or Responses tool group prevents every tool and pool replay (5611.4804ms)
✔ wire adapter preserves tool IDs but never passes Responses-private Items to Chat or another endpoint (0.2337ms)
✔ connection JSON import accepts explicit Responses protocol and continues to reject unknown protocols without reading client sessions (1230.8034ms)
✔ reservation uses actual known usage from the same route, never lowers estimates or rewrites original receipts (1.6681ms)
✔ reservation verification binds original Task/Run consumption and rejects changed usage or foreign provenance (0.4715ms)
✔ known high input calibrates the next request and stops before dispatch while preserving the completed read (2433.1102ms)
✔ one reported overrun stops legacy requests and keeps actual use and excess (1983.3313ms)
✔ one reported overrun stops pool requests and keeps actual use and excess (2819.3236ms)
✔ one reported overrun stops connection requests and keeps actual use and excess (2492.4876ms)
✔ reported usage below the limit prevents an unaffordable legacy retry before another dispatch (1999.818ms)
✔ reported usage below the limit prevents an unaffordable pool retry before another dispatch (3188.8902ms)
✔ native bridge seals calibrated refusal with original consumption and preserves versioned request verification (14678.9961ms)
✔ capability routing skips text-only members before acquiring a lease and records exact refusal in the original Run (2096.2784ms)
✔ all incompatible members reject with no network request, no lease, preserved Task and explicit non-dispatched ledger (1882.1025ms)
✔ explicit synthetic tool probe verifies only actual matching call, never executes it and loses verification on revision changes (1293.9178ms)
✔ plain text probe cannot verify tools; wrong tool nonce and client supplied evidence cannot claim verified support (1547.33ms)
✔ a disabled connection invalidates an in-flight probe even when toggle mutates its original registry object (1062.8668ms)
✔ selected connection owns the actual endpoint, credential, wire model and mismatch even when legacy is configured for GPT (3918.3035ms)
✔ budget-only API and demo/real selection preserve routes, reject injected endpoint fields and publish authoritative selection without keys (2898.388ms)
✔ another session cannot change default model, mode or budgets while a request is active (3509.7689ms)
✔ HTTP download preserves original malformed SSE bytes and artifact SHA instead of decoding invalid UTF8 (2306.3075ms)
✔ terminated SSE adapts empty optional tool arguments, retains the exact wire bytes and executes the validated batch once (3213.0203ms)
✔ reported SSE usage survives rejected required arguments "" without dispatching any tool (1802.1044ms)
✔ reported SSE usage survives rejected required arguments "{\"path\":" without dispatching any tool (1939.8046ms)
✔ reported SSE usage survives rejected required arguments "null" without dispatching any tool (2021.5647ms)
✔ a stream without DONE retains reported usage but cannot adapt or execute even an optional tool (1415.2064ms)
✔ a timeout after reasoning-only fragments creates one unknown receipt and never invents reported tokens (2238.2249ms)
✔ real Chat SSE carries UTF8 and split tool JSON; Runtime executes only after DONE, stores partial progress and original usage (3912.5031ms)
✔ Responses typed SSE validates original item identity, terminal data, raw usage and tool result pair (3332.6811ms)
✔ incomplete network stream retains unknown attempt and never automatically switches pool or dispatches fragment tools (2187.7844ms)
✔ stopping an active HTTP stream closes only original request and keeps partial text without executing tools (2097.4713ms)
✔ explicit streaming probe uses actual SSE despite unknown declaration and binds its own original attempt (1045.8503ms)
✔ coalesced initial SSE events publish the latest partial state even when the upstream stays idle (1463.6045ms)
✔ stream timeout closes the original HTTP response, preserves redacted capture and never retries (864.4765ms)
✔ stream idle timeout preserves incomplete tool arguments, records unknown use and never fails over (999.8465ms)
✔ SSE accepts CR-only final event boundaries and retains exact invalid UTF8 bytes without dispatch (0.8034ms)
▶ stream parser rejects truncation, finish length, malformed JSON, duplicate sequence, changed final item and oversized data
✔ reject 0 (0.2359ms)
✔ reject 1 (0.1577ms)
✔ reject 2 (0.1405ms)
✔ reject 3 (5.1929ms)
✔ reject 4 (0.3284ms)
✔ reject 5 (0.1975ms)
✔ stream parser rejects truncation, finish length, malformed JSON, duplicate sequence, changed final item and oversized data (6.7618ms)
✔ compiled host retains interrupted HTTP 200 and explicitly resumes after the recorded cooldown (8752.526ms)
✔ compiled host retains interrupted HTTP 503 and explicitly resumes after the recorded cooldown (9281.6574ms)
✔ owned IPv4 service works after a failed IPv6 candidate without a second model POST or global DNS change (25.684ms)
✔ environment HTTP proxy tunnels the owned request and NO_PROXY sends a separate direct request (10.6211ms)
✔ original connect timeout has a bounded nested cause and does not masquerade as response generation timeout (0.7106ms)
✔ POST accepted then reset is unknown, is charged once and never automatically retried (4.0045ms)
✔ partial HTTP bodies retain original request id and bytes, and an unknown pool attempt never switches model (2414.8937ms)
✔ response wait timeout after an accepted POST is unknown; cancellation closes the owned socket and permits explicit reuse (148.1235ms)
✔ hot upgrade adopts a retained virtual data root without reopening SQLite or changing rows, Provider or routing (1405.9648ms)
✔ failed adoption restores the original virtual data-root name and original open database (811.6903ms)
✔ hot upgrade refreshes retained model and importer methods while preserving credential and routing objects (915.3991ms)
✔ hot upgrade refuses active probes and restores retained model methods if installation fails (881.921ms)
ℹ tests 144
ℹ suites 0
ℹ pass 144
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 115256.5562
@@ -0,0 +1 @@
{"success":true,"version":"1.24.3","pid":36552,"projects":55,"sessions":71,"tables":{"events":12931,"metadata":6,"outbox":12931,"projects":55,"records":6086,"sessions":71,"sqlite_sequence":1},"runtimeRoot":"C:\\Users\\wangy\\AppData\\Local\\Programs\\AgentWorkbench\\versions\\Agent-Workbench-1.24.3-win-x64-20261103065830519","sourceCommit":"d62d03f93f0c1a94c3475bcc14385c4a3d9cde7a","hasKey":true,"active":[],"mode":"before","startedAt":"2026-11-03T11:44:21.703Z","finishedAt":"2026-11-03T11:44:30.410Z","realModelRequests":0,"scope":"Public API/full original row/export/artifact hashes; installer retains original running host and Provider. No credential extraction or task replay."}
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
@@ -0,0 +1,25 @@
> local-agent-workbench@1.24.3 build
> tsc --noEmit && vite build && node scripts/build-mobile.mjs && node scripts/build-server.mjs
vite v7.3.7 building client environment for production...
transforming...
✓ 1759 modules transformed.
rendering chunks...
computing gzip size...
dist-client/index.html 0.60 kB │ gzip: 0.41 kB
dist-client/.vite/manifest.json 0.65 kB │ gzip: 0.24 kB
dist-client/mobile.html 0.84 kB │ gzip: 0.47 kB
dist-client/assets/mobile-joLHctQw.css 12.39 kB │ gzip: 3.22 kB
dist-client/assets/desktop-AOdOAoy0.css 167.89 kB │ gzip: 31.71 kB
dist-client/assets/mobile-4zqcwTeD.js 27.19 kB │ gzip: 9.34 kB
dist-client/assets/approval-presentation-Chut4EFA.js 226.80 kB │ gzip: 71.65 kB
dist-client/assets/desktop-Dl2mfzfY.js 404.17 kB │ gzip: 114.60 kB
✓ built in 1.23s
Mobile shell built: 3 allowed assets; API and project data are never cached.
Server bundle built.
npm notice
npm notice New major version of npm available! 11.19.1 -> 12.2.1
npm notice Changelog: https://github.com/npm/cli/releases/tag/v12.2.1
npm notice To update run: npm install -g npm@12.2.1
npm notice
@@ -0,0 +1,59 @@
✔ identity distinguishes missing, invalid, mixed, exact and explicitly declared aliases without guessing (1.1899ms)
✔ chat-completions JSON rejects a different model before tool dispatch, retaining source and reported usage (2410.0353ms)
✔ chat-completions SSE rejects a different model before tool dispatch, retaining source and reported usage (1832.1492ms)
✔ responses JSON rejects a different model before tool dispatch, retaining source and reported usage (2114.8292ms)
✔ responses SSE rejects a different model before tool dispatch, retaining source and reported usage (1966.0531ms)
✔ record policy accepts missing identity but keeps it explicit, and strict declared mapping is frozen in the consumption (2.0347ms)
✔ mixed streamed identities cannot pass strict mode, and interrupted streams retain observed identifiers without authenticating them (1.5708ms)
✔ identity refusal never fails over to another pool member and raw response model cannot leak known credentials (2045.8864ms)
✔ strict refusal remains visible even when the same original response exhausts the budget (2006.2433ms)
✔ strict identity crosses the controlled engine bridge with its original source and zero tool dispatch (11411.0691ms)
✔ frozen direct connection policy and delivery reports preserve response identity without filling historical gaps (2647.6513ms)
✔ saved engine budgets verify all Chat/Responses route variants and reject a mismatched output allowance (1177.4117ms)
✔ owned CC Switch SQLite imports the selected TOML Responses wire_api and leaves unknown clients untouched (1132.0898ms)
✔ Responses executes one original tool through Runtime, carries encrypted Items, reports raw usage and seals exact wire bytes (3265.297ms)
✔ mixed-protocol pool keeps both actual variants and changes route on 429 without re-executing completed tools (3822.3256ms)
▶ Responses incomplete, failed, missing call identity and unsupported hosted outputs never dispatch a tool
✔ Reject {"body":{"id":"resp_owned","status":"incomplete","output":[{"type":"function_call","id":"fc_partial" (2178.3361ms)
✔ Reject {"body":{"id":"resp_owned","status":"failed","output":[{"type":"function_call","id":"fc_failed","cal (2051.5484ms)
✔ Reject {"body":{"id":"resp_owned","status":"completed","output":[{"type":"function_call","id":"fc_bad","cal (1604.752ms)
✔ Reject {"body":{"id":"resp_owned","status":"completed","output":[{"type":"computer_call","id":"not-supporte (1321.7878ms)
✔ Responses incomplete, failed, missing call identity and unsupported hosted outputs never dispatch a tool (7157.3552ms)
▶ invalid JSON in a nonstream Chat or Responses tool group prevents every tool and pool replay
✔ chat-completions (1415.8671ms)
✔ responses (1367.4864ms)
✔ invalid JSON in a nonstream Chat or Responses tool group prevents every tool and pool replay (2783.7407ms)
✔ wire adapter preserves tool IDs but never passes Responses-private Items to Chat or another endpoint (0.1895ms)
✔ connection JSON import accepts explicit Responses protocol and continues to reject unknown protocols without reading client sessions (499.1573ms)
✔ HTTP download preserves original malformed SSE bytes and artifact SHA instead of decoding invalid UTF8 (2316.2161ms)
✔ terminated SSE adapts empty optional tool arguments, retains the exact wire bytes and executes the validated batch once (3633.3617ms)
✔ reported SSE usage survives rejected required arguments "" without dispatching any tool (1492.5112ms)
✔ reported SSE usage survives rejected required arguments "{\"path\":" without dispatching any tool (1182.0086ms)
✔ reported SSE usage survives rejected required arguments "null" without dispatching any tool (1235.6693ms)
✔ a stream without DONE retains reported usage but cannot adapt or execute even an optional tool (1289.1869ms)
✔ a timeout after reasoning-only fragments creates one unknown receipt and never invents reported tokens (1518.3697ms)
✔ real Chat SSE carries UTF8 and split tool JSON; Runtime executes only after DONE, stores partial progress and original usage (2272.4199ms)
✔ Responses typed SSE validates original item identity, terminal data, raw usage and tool result pair (2026.1464ms)
✔ incomplete network stream retains unknown attempt and never automatically switches pool or dispatches fragment tools (1415.81ms)
✔ stopping an active HTTP stream closes only original request and keeps partial text without executing tools (1320.5973ms)
✔ explicit streaming probe uses actual SSE despite unknown declaration and binds its own original attempt (581.3611ms)
✔ coalesced initial SSE events publish the latest partial state even when the upstream stays idle (1513.5792ms)
✔ stream timeout closes the original HTTP response, preserves redacted capture and never retries (861.5278ms)
✔ stream idle timeout preserves incomplete tool arguments, records unknown use and never fails over (1027.5337ms)
✔ SSE accepts CR-only final event boundaries and retains exact invalid UTF8 bytes without dispatch (1.3354ms)
▶ stream parser rejects truncation, finish length, malformed JSON, duplicate sequence, changed final item and oversized data
✔ reject 0 (0.3689ms)
✔ reject 1 (0.2764ms)
✔ reject 2 (0.2227ms)
✔ reject 3 (5.2073ms)
✔ reject 4 (0.1789ms)
✔ reject 5 (0.1875ms)
✔ stream parser rejects truncation, finish length, malformed JSON, duplicate sequence, changed final item and oversized data (7.2245ms)
ℹ tests 48
ℹ suites 0
ℹ pass 48
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 43995.5843
@@ -0,0 +1,9 @@
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\messages-ui-m-7ec38a5c-68d3-4290-be14-56d0e969a264
PASS root and /v1 preview agree, full endpoint rejects with input retained; save makes no model request
PASS saved protocol, auth and selected connection survive reload and remain operable at 390px
PASS native settings reject other-protocol fields and keep a single hard max_tokens output field
PASS actual local JSON text/tool and native SSE capability probes use the saved protocol and auth
PASS compiled native tool roundtrip completes with original call ID, cache-inclusive usage and observed request/response URLs
PASS strict mismatch is visible with original GPT identifier and one request, no retry or tool dispatch
PASS 429 preserves native type, raw message, request ID, URL and inputs without automatic replay
{"success":true,"checks":7,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-ui-m-7ec38a5c-68d3-4290-be14-56d0e969a264"}
@@ -0,0 +1,18 @@
{
"version": "1.24.4",
"sourceCommit": "65ddee16b0798b72e9e5780250e2f4f2b5d6dda9",
"candidate": "Agent-Workbench-1.24.4-win-x64-20261103105910315",
"package": "complete",
"archives": "all three CRC, manifest and directory bytes passed",
"independentBuild": "549 extracted inputs, offline install, 22 compiled files all identical",
"productFreeze": "544 tracked product inputs and 22 compiled files",
"portable": "passed",
"portableMessagesUI": "7 groups passed, realModelRequests=0",
"probeDriver": "three isolated scenarios passed; match, strict mismatch, raw 429; each one request and zero tools; inspector closed",
"probeDriverFailures": "two preserved QA-only event-count mistakes before using distinct wireAttemptId; original logs and scripts retained",
"isolatedWindowsInstall": "running",
"formal": "still original 1.24.3/PID36552/Provider; no credential extraction; pending upgrade",
"realPreflight": "pending; never replay old J/E/H; 1 request/256 output/16000 total/0 tools/0 retries/0 pool switch",
"publicConfigObservation": "existing apikey credentialRevision2 present; its older saved DPAPI block was already source_changed before M changes",
"documentation": "tracked reports and plans deferred until formal installer clean-source checks complete; this receipt preserves intermediate findings"
}
@@ -0,0 +1,13 @@
PASS actual clean user-level installation
PASS actual installer rejects an open attested desktop window
PASS actual update rejects a pending owned model request
PASS actual installer rejects a persistent original task approval
PASS actual manifest fault installer preserves original pointer, PID and all data
PASS actual pointer-published fault rolls original host and pointer back
PASS actual update retains original PID, instance, DPAPI ciphertext and every persistent byte
PASS installed native bootstrap verifies and opens the registered release
PASS native desktop remains registered before explicit owned cleanup
PASS real host restart restores the owned saved credential and actual HTTP authentication
PASS actual installed current results and saved HTML user paths
PASS actual uninstall closes only the idle owned host and preserves user data
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\desktop-install-5a966b14-29e6-4fe6-bebc-3ed92bd8b93c
@@ -0,0 +1,33 @@
PASS actual clean user-level installation
PASS actual installer rejects an open attested desktop window
PASS actual update rejects a pending owned model request
PASS actual installer rejects a persistent original task approval
PASS actual manifest fault installer preserves original pointer, PID and all data
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\desktop-install-f09eea3e-1ef6-410a-8322-ca66f6deba0f
node:internal/modules/run_main:107
triggerUncaughtException(
^
AssertionError [ERR_ASSERTION]: The input did not match the regular expression /owned-post-pointer-fault/. Input:
'Error: This migration accepts a supported existing Workbench.'
at process.processTicksAndRejections (node:internal/process/task_queues:104:5)
at async file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/scripts/verify-desktop-install.mjs:119:3 {
generatedMessage: true,
code: 'ERR_ASSERTION',
actual: Error: This migration accepts a supported existing Workbench.
at upgradeRunning (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/scripts/upgrade-running.mjs:67:424)
at process.processTicksAndRejections (node:internal/process/task_queues:104:5)
at async file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/core/desktop/install-manager.mjs:196:11
at async file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/core/desktop/install-manager.mjs:78:22
at async InstallationManager.lock (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/core/desktop/install-manager.mjs:151:18)
at async waitForActual (node:assert:646:5)
at async strict.rejects (node:assert:769:25)
at async file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/scripts/verify-desktop-install.mjs:119:3,
expected: /owned-post-pointer-fault/,
operator: 'rejects',
diff: 'simple'
}
Node.js v24.22.1
@@ -0,0 +1 @@
{"passed":true,"checks":[{"table":"events","original":12931,"current":12953,"exact":12931,"declared":0},{"table":"metadata","original":6,"current":6,"exact":4,"declared":2},{"table":"outbox","original":12931,"current":12953,"exact":12931,"declared":0},{"table":"projects","original":55,"current":56,"exact":0,"declared":55},{"table":"records","original":6086,"current":6094,"exact":6086,"declared":0},{"table":"sessions","original":71,"current":72,"exact":71,"declared":0},{"table":"sqlite_sequence","original":1,"current":1,"exact":0,"declared":1}],"originalExports":71,"originalArtifacts":3570,"newRows":54}
@@ -0,0 +1,2 @@
{"stage":"single-preflight","requestUrl":"https://api.apikey.fan/v1/messages","model":"claude-opus-5-5","tools":0,"maxRequests":1}
{"completedAt":"2026-11-03T12:03:16.397Z","version":"1.24.4","pid":36552,"connectionId":"09b0293b-faad-45c3-832d-7c8b615813aa","sessionId":"f7d1ec3d-aa41-4f4e-ab6d-f4df487b4d76","taskId":"15d22836-27a0-422c-a0a3-d0ed1752d1f4","runId":"b175e0e0-1472-42af-99dd-bda5d1edcc0b","protocol":"anthropic-messages","messagesAuth":"bearer","requestUrl":"https://api.apikey.fan/v1/messages","responseUrl":"https://api.apikey.fan/v1/messages","requestedModel":"claude-opus-5-5","reportedModels":["gpt-6.1-sol"],"identityStatus":"mismatch","identityPassed":false,"sessionStatus":"paused","stopReason":"model_identity","requests":1,"toolDefinitions":0,"actualTools":0,"usage":{"prompt_tokens":192,"completion_tokens":29,"total_tokens":221,"prompt_tokens_details":{"cached_tokens":0,"cache_creation_tokens":0},"raw":{"input_tokens":192,"output_tokens":29,"cache_creation_input_tokens":0,"cache_read_input_tokens":0},"protocol":"anthropic-messages"},"usageStatus":"reported","runUsage":{"requests":1,"inputTokens":192,"outputTokens":29,"totalTokens":221,"unknownRequests":0,"estimatedUnknownTokens":0},"captures":[{"kind":"request","file":"request.json","id":"8f508263-0025-4553-a194-1daff505fc22","bytes":440,"sha256":"5420746a315e7226d1636d6229bdb4545b2089c3f3e36c2019bfce32e250ef40"},{"kind":"response","file":"response.sse","id":"eeb6c2ef-c9aa-4202-ab19-dbc19877d0bc","bytes":3214,"sha256":"403a817ff9c9b4d4f7bd0094eac561cd358a541398a4875c9167a08ece1482c4"}],"noRetry":true,"noPoolSwitch":true,"modelWeightsAuthenticated":false,"realEngineeringDeliveryTested":false,"originalPurgeBenefitTested":false,"inspectorClosed":true,"defaultSelectionRetained":true,"credentialRevisionRetained":true}
@@ -0,0 +1,84 @@
{
"portableRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103112054864",
"sourceRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-source-20261103112054864",
"evidenceRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-evidence-20261103112054864",
"portable": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103112054864.zip",
"bytes": 242709419,
"sha256": "31983cb4258ebcf517ee66f5da2b4e0464cc0e04620341a2491b38b5d56ce329"
},
"source": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-source-20261103112054864.zip",
"bytes": 894781411,
"sha256": "20c6c851480abaec78de0860885f41370e27773863ae96d0dac97f0e437d51d0"
},
"evidence": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-evidence-20261103112054864.zip",
"bytes": 653648820,
"sha256": "3b3ea76e98739b657fda91bdb1d95e60f451346f9d6453125a53aa6dc7ae4ef2"
},
"documentation": {
"sourceCommit": "d62d03f93f0c1a94c3475bcc14385c4a3d9cde7a",
"originalFiles": 18877,
"originalBytes": 2026294838,
"shippedFiles": 141,
"shippedBytes": 1091320,
"evidenceManifestSha256": "f3dfb490d57fd60e7891ed347a31bf4afeab5649798138a370c11bd4c32fd937"
},
"installer": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103112054864-Setup.exe",
"bytes": 213126171,
"sha256": "4012a7914c4e0f44f93c56d6603c9c2000b3090971fee2a688855de3c8c94960",
"compiler": {
"version": "6.3.1",
"sha256": "3b22d36fc7e7c421d8caf780008ecdb42b14b3f0371b9bafe3cad64e6648d9cf"
},
"source": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\installer-build-f7b3301c-b38c-4143-aaee-d91718c3c75a\\installer.iss",
"version": "1.24.4",
"candidateDirectory": "Agent-Workbench-1.24.4-win-x64-20261103112054864",
"qaFixture": null
},
"desktopRuntime": {
"version": "45.2.0",
"sha256": "010ea51bfb528c8143513439aa93fb40754403daa2023f9decd4517da71706ab",
"bytes": 249810432,
"executable": "desktop/AgentWorkbench.exe",
"scope": "Fixed original Windows Electron runtime, full license texts, isolated application loader and native bootstrap; original executable hash unchanged."
},
"nativeLauncher": {
"bytes": 12288,
"sha256": "e3d31cb7ce5009d3ca5a96d04dc16b1add0bd1f5cde7e1304461de6713d37463",
"inputs": {
"compiler": "46809206887326d2d24db1eff1f3064de972c3451abe766b49111450a5e08e00",
"source": "1c8e7d7ee5c383897bec48e9aadb8051b47c09686f74bbf34e0b6b695cd01ae5",
"icon": "f1879074e45f01c480ad9bf50b818637cc60ad56526056c886f808a4812d7b0c"
},
"compiler": "Windows .NET Framework C# compiler; source-bound cached original output, no deterministic-build claim."
},
"nodeVersion": "v24.22.1",
"reaRuntime": {
"version": "6.8.0",
"protocolVersion": "2025-03-26",
"packages": 149,
"files": 8413,
"bytes": 142395402,
"lockSha256": "687fb3fa3089a12f239dd5e87d9fa07df55a87df4935931d495a92600a1b0f6e"
},
"portableFiles": 10936,
"licenseTexts": 135,
"browserRuntime": {
"playwrightVersion": "1.64.1",
"packages": [
{
"name": "playwright",
"version": "1.64.1",
"files": 60
},
{
"name": "playwright-core",
"version": "1.64.1",
"files": 115
}
]
}
}
@@ -0,0 +1,84 @@
{
"portableRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103105910315",
"sourceRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-source-20261103105910315",
"evidenceRoot": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-evidence-20261103105910315",
"portable": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103105910315.zip",
"bytes": 242709255,
"sha256": "78256102444dd5030f8767e66d54d776a8a788c6eca02d804815fa24c2db2c53"
},
"source": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-source-20261103105910315.zip",
"bytes": 894780480,
"sha256": "22854813be9920547d88ca71313af73889a654e817fd3613747f588c33429f9c"
},
"evidence": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-evidence-20261103105910315.zip",
"bytes": 653648819,
"sha256": "26ab3e926b39d70975b528980142eaba2fec10b941b1ef6484b5be08d656ef2e"
},
"documentation": {
"sourceCommit": "65ddee16b0798b72e9e5780250e2f4f2b5d6dda9",
"originalFiles": 18877,
"originalBytes": 2026294838,
"shippedFiles": 141,
"shippedBytes": 1091320,
"evidenceManifestSha256": "69075f59e0b92b6408cd6160d12ac1aa5a7288adbfcc040b5524f9a4c2e483b8"
},
"installer": {
"path": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-win-x64-20261103105910315-Setup.exe",
"bytes": 213126189,
"sha256": "c60b536feec258821759a447a559fd625152f5c0bb60b07cb692a648637a6ce3",
"compiler": {
"version": "6.3.1",
"sha256": "3b22d36fc7e7c421d8caf780008ecdb42b14b3f0371b9bafe3cad64e6648d9cf"
},
"source": "D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\installer-build-e2ad6e37-30e8-49df-8486-43eba522e8a8\\installer.iss",
"version": "1.24.4",
"candidateDirectory": "Agent-Workbench-1.24.4-win-x64-20261103105910315",
"qaFixture": null
},
"desktopRuntime": {
"version": "45.2.0",
"sha256": "010ea51bfb528c8143513439aa93fb40754403daa2023f9decd4517da71706ab",
"bytes": 249810432,
"executable": "desktop/AgentWorkbench.exe",
"scope": "Fixed original Windows Electron runtime, full license texts, isolated application loader and native bootstrap; original executable hash unchanged."
},
"nativeLauncher": {
"bytes": 12288,
"sha256": "e3d31cb7ce5009d3ca5a96d04dc16b1add0bd1f5cde7e1304461de6713d37463",
"inputs": {
"compiler": "46809206887326d2d24db1eff1f3064de972c3451abe766b49111450a5e08e00",
"source": "1c8e7d7ee5c383897bec48e9aadb8051b47c09686f74bbf34e0b6b695cd01ae5",
"icon": "f1879074e45f01c480ad9bf50b818637cc60ad56526056c886f808a4812d7b0c"
},
"compiler": "Windows .NET Framework C# compiler; source-bound cached original output, no deterministic-build claim."
},
"nodeVersion": "v24.22.1",
"reaRuntime": {
"version": "6.8.0",
"protocolVersion": "2025-03-26",
"packages": 149,
"files": 8413,
"bytes": 142395402,
"lockSha256": "687fb3fa3089a12f239dd5e87d9fa07df55a87df4935931d495a92600a1b0f6e"
},
"portableFiles": 10936,
"licenseTexts": 135,
"browserRuntime": {
"playwrightVersion": "1.64.1",
"packages": [
{
"name": "playwright",
"version": "1.64.1",
"files": 60
},
{
"name": "playwright-core",
"version": "1.64.1",
"files": 115
}
]
}
}
@@ -0,0 +1,3 @@
PASS offline-install
PASS build
{"success":true,"root":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\packaged-build-6947e11b-e3cc-4f3f-afdc-dba4c424264c","output":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\packaged-build-6947e11b-e3cc-4f3f-afdc-dba4c424264c\\result.json","inputCount":549,"compiledFiles":22,"mismatches":0}
@@ -0,0 +1,3 @@
PASS offline-install
PASS build
{"success":true,"root":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\packaged-build-83597e7e-d7fd-4514-938b-de21e0da3ac5","output":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\packaged-build-83597e7e-d7fd-4514-938b-de21e0da3ac5\\result.json","inputCount":549,"compiledFiles":22,"mismatches":0}
File diff suppressed because one or more lines are too long
@@ -0,0 +1,9 @@
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\messages-ui-m-517c8fd8-11db-4079-a5f4-29a78662aeb8
PASS root and /v1 preview agree, full endpoint rejects with input retained; save makes no model request
PASS saved protocol, auth and selected connection survive reload and remain operable at 390px
PASS native settings reject other-protocol fields and keep a single hard max_tokens output field
PASS actual local JSON text/tool and native SSE capability probes use the saved protocol and auth
PASS compiled native tool roundtrip completes with original call ID, cache-inclusive usage and observed request/response URLs
PASS strict mismatch is visible with original GPT identifier and one request, no retry or tool dispatch
PASS 429 preserves native type, raw message, request ID, URL and inputs without automatic replay
{"success":true,"checks":7,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-ui-m-517c8fd8-11db-4079-a5f4-29a78662aeb8"}
@@ -0,0 +1,9 @@
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\messages-ui-m-1fc6d898-8431-482d-bbe4-56241289798a
PASS root and /v1 preview agree, full endpoint rejects with input retained; save makes no model request
PASS saved protocol, auth and selected connection survive reload and remain operable at 390px
PASS native settings reject other-protocol fields and keep a single hard max_tokens output field
PASS actual local JSON text/tool and native SSE capability probes use the saved protocol and auth
PASS compiled native tool roundtrip completes with original call ID, cache-inclusive usage and observed request/response URLs
PASS strict mismatch is visible with original GPT identifier and one request, no retry or tool dispatch
PASS 429 preserves native type, raw message, request ID, URL and inputs without automatic replay
{"success":true,"checks":7,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-ui-m-1fc6d898-8431-482d-bbe4-56241289798a"}
File diff suppressed because one or more lines are too long
@@ -0,0 +1,12 @@
file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/.qa/messages-m-inspector.mjs:48
if (result.exceptionDetails) throw Error(result.exceptionDetails.exception?.description ?? result.exceptionDetails.text);
^
Error: Error: Single-request limit exceeded.
at ensure (<anonymous>:2:58)
at nativeMessagesProbe (<anonymous>:78:3)
at invokeProbe (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/.qa/messages-m-inspector.mjs:48:40)
at process.processTicksAndRejections (node:internal/process/task_queues:104:5)
at async file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/.qa/messages-m-probe-fixture.mjs:43:20
Node.js v24.22.1
@@ -0,0 +1,4 @@
{"mode":"match","passed":true,"requests":1,"actualTools":0,"status":"completed","inspectorClosed":true}
{"mode":"mismatch","passed":true,"requests":1,"actualTools":0,"status":"paused","inspectorClosed":true}
{"mode":"429","passed":true,"requests":1,"actualTools":0,"status":"failed","inspectorClosed":true}
{"passed":true,"root":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-m\\probe-fixture-ee6cedb6-d0cd-4201-9df9-1ace3255a6ac"}
@@ -0,0 +1,12 @@
file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/.qa/messages-m-inspector.mjs:48
if (result.exceptionDetails) throw Error(result.exceptionDetails.exception?.description ?? result.exceptionDetails.text);
^
Error: Error: Single-request limit exceeded.
at ensure (<anonymous>:2:58)
at nativeMessagesProbe (<anonymous>:78:3)
at invokeProbe (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/.qa/messages-m-inspector.mjs:48:40)
at process.processTicksAndRejections (node:internal/process/task_queues:104:5)
at async file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/.qa/messages-m-probe-fixture.mjs:43:20
Node.js v24.22.1
@@ -0,0 +1,69 @@
✔ a Windows virtual data-directory name binds new storage and credentials to its actual physical root (2006.0181ms)
✔ physical data-root adoption still rejects explicit junctions at the directory or any ancestor (956.7685ms)
✔ actual Windows CurrentUser DPAPI roundtrips owned UTF-8 and rejects damaged bytes and wrong entropy (2363.9819ms)
✔ only explicitly supplied keys persist; new provider restores original revisions and public views contain no key (2518.5943ms)
✔ registry and cipher commit failure restores original routing, keys, block and every database row (1920.6749ms)
✔ a two-key import with a protection failure rolls back all registry and cipher entries and retains its preview (1924.943ms)
✔ connection revision changes retain the original saved block and report a source mismatch on restart (2239.3686ms)
✔ disable skips real restart loading, explicit enable restores, delete preserves memory and forget removes both (3013.864ms)
✔ corrupt saved block and another SID do not load; failures are visible without returning ciphertext or plaintext (2507.0275ms)
✔ legacy persistence requires a new explicit key and binds config and credential revisions; stale requests cannot save (2630.4426ms)
✔ owned plaintext keys never reach SQLite pages, WAL, config or public execution exports (1343.3042ms)
✔ model registry keeps original credentials/endpoint intact, rejects stale edits, and forgets only volatile keys on restart (1294.0386ms)
✔ config import previews exact revisions, keeps secrets volatile, syncs only when enabled and preserves manual edits on conflicts (762.952ms)
✔ real HTTP pool fails over 429 to another model and never repeats completed tools; attempts seal actual request bytes and Run identity (2868.5631ms)
✔ pool rotation and weights select members on new Runs; disabled and missing credentials are excluded without requests (4292.5961ms)
✔ 401 blocks the credential revision; updating the key restores availability without leaking error echoes (1607.8489ms)
✔ queued request cancellation and in-flight disable release only owned leases and do not dispatch another HTTP request (704.8182ms)
✔ compiled model-center HTTP API persists public profiles, blocks forged origins, and local operations never invoke a model (2261.4197ms)
✔ model registry disk failures restore routing, pool profiles and credential references without partial updates (994.5932ms)
✔ model selection and saved configuration commit together and failures preserve the original mode and endpoint (649.3802ms)
✔ configuration imports roll back the entire batch and retain one usable preview after validation or commit failure (741.326ms)
✔ HTTP quota retains native code/type/param/request id, bounded original and allowlisted headers, never repeats that connection (1628.9828ms)
✔ Retry-After above 90 seconds remains exact and a short local budget cannot trigger an early retry (1.7788ms)
✔ Retry-After date, expired date, huge value and invalid header retain explicit semantics (1.1084ms)
✔ unclassified 429 remains unspecified and every bounded actual retry retains its own original error (1372.273ms)
✔ oversized HTTP error saves exactly the bounded prefix and labels its incomplete range (1524.2275ms)
✔ Chat SSE error preserves upstream fields, original failed event, partial tool and reported usage without executing or retrying (1555.7782ms)
✔ Responses failed SSE retains original event/response identity and usage (1365.7143ms)
✔ Responses incomplete SSE retains original event/response identity and usage (2372.8577ms)
✔ unsupported successful Responses output retains original JSON and usage with a local protocol origin (2192.4296ms)
✔ HTTP 200 native error is still an upstream failure rather than a fabricated successful answer (0.8ms)
✔ JSON-escaped credential echoes are redacted in the original preview and capture as well as parsed fields (0.7165ms)
✔ quota failover uses another independent pool member and never replays a failed stream (2866.6846ms)
✔ native parameters are isolated by pool member, preserve Responses state and bind the original output reservation (0.3202ms)
✔ unsupported native fields and invalid values reject before dispatch instead of silently dropping or overriding constraints (0.328ms)
✔ native settings for legacy and registered profiles persist without changing credentials or old records and reject stale edits (1070.7599ms)
✔ compiled native settings API saves/restores exact parameters and rejects reserved fields without mutating the profile (2439.4011ms)
✔ legacy connection test sends its saved native fields and selected output limit field without changing the configuration (2281.012ms)
✔ reservation uses actual known usage from the same route, never lowers estimates or rewrites original receipts (1.6179ms)
✔ reservation verification binds original Task/Run consumption and rejects changed usage or foreign provenance (0.4752ms)
✔ known high input calibrates the next request and stops before dispatch while preserving the completed read (2743.551ms)
✔ one reported overrun stops legacy requests and keeps actual use and excess (1813.1795ms)
✔ one reported overrun stops pool requests and keeps actual use and excess (2469.8499ms)
✔ one reported overrun stops connection requests and keeps actual use and excess (2080.7179ms)
✔ reported usage below the limit prevents an unaffordable legacy retry before another dispatch (1242.4605ms)
✔ reported usage below the limit prevents an unaffordable pool retry before another dispatch (1835.2875ms)
✔ native bridge seals calibrated refusal with original consumption and preserves versioned request verification (12172.4228ms)
✔ capability routing skips text-only members before acquiring a lease and records exact refusal in the original Run (1738.0465ms)
✔ all incompatible members reject with no network request, no lease, preserved Task and explicit non-dispatched ledger (891.5137ms)
✔ explicit synthetic tool probe verifies only actual matching call, never executes it and loses verification on revision changes (667.351ms)
✔ plain text probe cannot verify tools; wrong tool nonce and client supplied evidence cannot claim verified support (735.7068ms)
✔ a disabled connection invalidates an in-flight probe even when toggle mutates its original registry object (685.5594ms)
✔ selected connection owns the actual endpoint, credential, wire model and mismatch even when legacy is configured for GPT (2163.8534ms)
✔ budget-only API and demo/real selection preserve routes, reject injected endpoint fields and publish authoritative selection without keys (2632.9035ms)
✔ another session cannot change default model, mode or budgets while a request is active (3900.9296ms)
✔ owned IPv4 service works after a failed IPv6 candidate without a second model POST or global DNS change (25.9974ms)
✔ environment HTTP proxy tunnels the owned request and NO_PROXY sends a separate direct request (11.066ms)
✔ original connect timeout has a bounded nested cause and does not masquerade as response generation timeout (0.82ms)
✔ POST accepted then reset is unknown, is charged once and never automatically retried (4.2966ms)
✔ partial HTTP bodies retain original request id and bytes, and an unknown pool attempt never switches model (2128.0014ms)
✔ response wait timeout after an accepted POST is unknown; cancellation closes the owned socket and permits explicit reuse (146.9489ms)
ℹ tests 61
ℹ suites 0
ℹ pass 61
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 51021.9798
@@ -0,0 +1,18 @@
node:fs:621
return binding.open(
^
Error: ENOENT: no such file or directory, open 'D:\Ai-project\REA逆向\Agent-Workbench\releases\Agent-Workbench-1.24.4-source-20261103112054864\dist-client\.vite\manifest.json'
at Object.openSync (node:fs:621:18)
at Object.readFileSync (node:fs:487:35)
at file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/.qa/messages-m-seal.mjs:14:77
at ModuleJob.run (node:internal/modules/esm/module_job:561:25)
at async node:internal/modules/esm/loader:647:26
at async asyncRunEntryPointWithESMLoader (node:internal/modules/run_main:101:5) {
errno: -4058,
code: 'ENOENT',
syscall: 'open',
path: 'D:\\Ai-project\\REA逆向\\Agent-Workbench\\releases\\Agent-Workbench-1.24.4-source-20261103112054864\\dist-client\\.vite\\manifest.json'
}
Node.js v24.22.1
@@ -0,0 +1,34 @@
✔ Messages builds native system/tool history, bounds images, and seals a distinct mixed-pool variant (6.533ms)
✔ Messages URL and native parameter contracts preserve other protocol paths and hard output budget (0.686ms)
✔ native usage includes both cache buckets once, retains raw and rejects invalid counters (0.2148ms)
✔ real HTTP Messages JSON executes one tool, retains opaque history, actual URL and cache accounting (2131.0459ms)
✔ real HTTP Messages SSE executes one tool, retains opaque history, actual URL and cache accounting (2072.9066ms)
✔ Messages api-key authentication is explicit; mismatched model remains blocked with original usage (4.9657ms)
✔ Messages bearer authentication is explicit; mismatched model remains blocked with original usage (4.1735ms)
✔ Messages redirect is not followed, either credential stays at the original server (3.6458ms)
✔ strict native SSE reads the nested start model and retains mismatch source without retries (2.9781ms)
✔ a closed native tool block cannot execute before message_stop, then executes exactly once (2344.3154ms)
✔ mixed Chat/Messages pool seals both variants and preserves one completed tool on failover (2432.9891ms)
✔ native JSON rejects incomplete stops, duplicate tools and hosted content (0.4488ms)
▶ strict SSE validates all blocks, terminal, event names, UTF8 and tool JSON before dispatch
✔ reject native stream 0 (1.5562ms)
✔ reject native stream 1 (0.2573ms)
✔ reject native stream 2 (0.2434ms)
✔ reject native stream 3 (0.2015ms)
✔ reject native stream 4 (0.1671ms)
✔ reject native stream 5 (0.1992ms)
✔ reject native stream 6 (0.1361ms)
✔ reject native stream 7 (0.1296ms)
✔ reject native stream 8 (0.5456ms)
✔ strict SSE validates all blocks, terminal, event names, UTF8 and tool JSON before dispatch (4.2376ms)
✔ upstream SSE error retains original type/request ID and initial usage, never retries (4.2842ms)
✔ stop cancels the owned stream and retains partial consumption without running tool or replaying (59.0864ms)
✔ saved Messages import preserves auth choice, rejects endpoint duplication, and keeps source file untouched (540.091ms)
ℹ tests 25
ℹ suites 0
ℹ pass 25
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 9805.3693
@@ -0,0 +1,54 @@
✔ Messages builds native system/tool history, bounds images, and seals a distinct mixed-pool variant (6.7439ms)
✔ Messages URL and native parameter contracts preserve other protocol paths and hard output budget (0.5237ms)
✔ native usage includes both cache buckets once, retains raw and rejects invalid counters (0.1551ms)
✔ real HTTP Messages JSON executes one tool, retains opaque history, actual URL and cache accounting (2052.8858ms)
✔ real HTTP Messages SSE executes one tool, retains opaque history, actual URL and cache accounting (2199.8092ms)
✔ Messages api-key authentication is explicit; mismatched model remains blocked with original usage (5.337ms)
✔ Messages bearer authentication is explicit; mismatched model remains blocked with original usage (4.5149ms)
✖ Messages redirect is not followed, either credential stays at the original server (4.2525ms)
✔ mixed Chat/Messages pool seals both variants and preserves one completed tool on failover (2463.7458ms)
✔ native JSON rejects incomplete stops, duplicate tools and hosted content (0.2863ms)
▶ strict SSE validates all blocks, terminal, event names, UTF8 and tool JSON before dispatch
✔ reject native stream 0 (1.1281ms)
✔ reject native stream 1 (0.2538ms)
✔ reject native stream 2 (0.1874ms)
✔ reject native stream 3 (0.169ms)
✔ reject native stream 4 (0.1337ms)
✔ reject native stream 5 (0.1425ms)
✔ reject native stream 6 (0.1494ms)
✔ reject native stream 7 (0.1256ms)
✔ reject native stream 8 (0.1693ms)
✔ strict SSE validates all blocks, terminal, event names, UTF8 and tool JSON before dispatch (3.0188ms)
✔ upstream SSE error retains original type/request ID and initial usage, never retries (4.7391ms)
✔ stop cancels the owned stream and retains partial consumption without running tool or replaying (67.1193ms)
✔ saved Messages import preserves auth choice, rejects endpoint duplication, and keeps source file untouched (445.5612ms)
ℹ tests 23
ℹ suites 0
ℹ pass 22
ℹ fail 1
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 7452.8426
✖ failing tests:
test at tests\anthropic-messages.test.mjs:143:1
✖ Messages redirect is not followed, either credential stays at the original server (4.2525ms)
AssertionError [ERR_ASSERTION]: Expected values to be strictly equal:
+ actual - expected
+ undefined
- 'http://127.0.0.1:49989/v1/messages'
at TestContext.<anonymous> (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/tests/anthropic-messages.test.mjs:148:10)
at process.processTicksAndRejections (node:internal/process/task_queues:104:5)
at async Test.run (node:internal/test_runner/test:1409:7)
at async Test.processPendingSubtests (node:internal/test_runner/test:974:7) {
generatedMessage: true,
code: 'ERR_ASSERTION',
actual: undefined,
expected: 'http://127.0.0.1:49989/v1/messages',
operator: 'strictEqual',
diff: 'simple'
}
@@ -0,0 +1,32 @@
✔ Messages builds native system/tool history, bounds images, and seals a distinct mixed-pool variant (6.536ms)
✔ Messages URL and native parameter contracts preserve other protocol paths and hard output budget (0.5297ms)
✔ native usage includes both cache buckets once, retains raw and rejects invalid counters (0.1594ms)
✔ real HTTP Messages JSON executes one tool, retains opaque history, actual URL and cache accounting (1781.7617ms)
✔ real HTTP Messages SSE executes one tool, retains opaque history, actual URL and cache accounting (1979.8322ms)
✔ Messages api-key authentication is explicit; mismatched model remains blocked with original usage (5.0219ms)
✔ Messages bearer authentication is explicit; mismatched model remains blocked with original usage (4.5076ms)
✔ Messages redirect is not followed, either credential stays at the original server (4.02ms)
✔ mixed Chat/Messages pool seals both variants and preserves one completed tool on failover (2540.0846ms)
✔ native JSON rejects incomplete stops, duplicate tools and hosted content (0.3453ms)
▶ strict SSE validates all blocks, terminal, event names, UTF8 and tool JSON before dispatch
✔ reject native stream 0 (1.0822ms)
✔ reject native stream 1 (0.4539ms)
✔ reject native stream 2 (0.1931ms)
✔ reject native stream 3 (0.2031ms)
✔ reject native stream 4 (0.2173ms)
✔ reject native stream 5 (0.1571ms)
✔ reject native stream 6 (0.1375ms)
✔ reject native stream 7 (0.1181ms)
✔ reject native stream 8 (0.1902ms)
✔ strict SSE validates all blocks, terminal, event names, UTF8 and tool JSON before dispatch (3.3462ms)
✔ upstream SSE error retains original type/request ID and initial usage, never retries (5.4571ms)
✔ stop cancels the owned stream and retains partial consumption without running tool or replaying (58.6133ms)
✔ saved Messages import preserves auth choice, rejects endpoint duplication, and keeps source file untouched (497.7334ms)
ℹ tests 23
ℹ suites 0
ℹ pass 23
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 7082.2029
@@ -0,0 +1,9 @@
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\messages-ui-m-9054fbf2-0271-41b7-b748-f53a05419ee9
PASS root and /v1 preview agree, full endpoint rejects with input retained; save makes no model request
PASS saved protocol, auth and selected connection survive reload and remain operable at 390px
PASS native settings reject other-protocol fields and keep a single hard max_tokens output field
PASS actual local JSON text/tool and native SSE capability probes use the saved protocol and auth
PASS compiled native tool roundtrip completes with original call ID, cache-inclusive usage and observed request/response URLs
PASS strict mismatch is visible with original GPT identifier and one request, no retry or tool dispatch
PASS 429 preserves native type, raw message, request ID, URL and inputs without automatic replay
{"success":true,"checks":7,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-ui-m-9054fbf2-0271-41b7-b748-f53a05419ee9"}
@@ -0,0 +1,9 @@
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\messages-ui-m-41953afc-ebd4-4a83-ae2b-f543df3a66d2
PASS root and /v1 preview agree, full endpoint rejects with input retained; save makes no model request
PASS saved protocol, auth and selected connection survive reload and remain operable at 390px
PASS native settings reject other-protocol fields and keep a single hard max_tokens output field
PASS actual local JSON text/tool and native SSE capability probes use the saved protocol and auth
PASS compiled native tool roundtrip completes with original call ID, cache-inclusive usage and observed request/response URLs
PASS strict mismatch is visible with original GPT identifier and one request, no retry or tool dispatch
PASS 429 preserves native type, raw message, request ID, URL and inputs without automatic replay
{"success":true,"checks":7,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-ui-m-41953afc-ebd4-4a83-ae2b-f543df3a66d2"}
@@ -0,0 +1,5 @@
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\model-selection-k-ui-regression-86916c34-8568-4e02-beaf-26c5400969e0
{"script":"verify-model-center-ui","code":0}
{"script":"verify-model-protocols-ui","code":0}
{"script":"verify-model-identity-ui","code":0}
{"script":"verify-model-selection-ui","code":0}
@@ -0,0 +1,9 @@
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\messages-ui-m-65bee3ef-4167-4a79-9d3f-6dc9a4fbd409
PASS root and /v1 preview agree, full endpoint rejects with input retained; save makes no model request
PASS saved protocol, auth and selected connection survive reload and remain operable at 390px
PASS native settings reject other-protocol fields and keep a single hard max_tokens output field
PASS actual local JSON text/tool and native SSE capability probes use the saved protocol and auth
PASS compiled native tool roundtrip completes with original call ID, cache-inclusive usage and observed request/response URLs
PASS strict mismatch is visible with original GPT identifier and one request, no retry or tool dispatch
PASS 429 preserves native type, raw message, request ID, URL and inputs without automatic replay
{"success":true,"checks":7,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-ui-m-65bee3ef-4167-4a79-9d3f-6dc9a4fbd409"}
@@ -0,0 +1,9 @@
EVIDENCE D:\Ai-project\REA逆向\Agent-Workbench\.qa\messages-ui-m-fae17b5c-0946-44b2-b655-f17521ae9af4
PASS root and /v1 preview agree, full endpoint rejects with input retained; save makes no model request
PASS saved protocol, auth and selected connection survive reload and remain operable at 390px
PASS native settings reject other-protocol fields and keep a single hard max_tokens output field
PASS actual local JSON text/tool and native SSE capability probes use the saved protocol and auth
PASS compiled native tool roundtrip completes with original call ID, cache-inclusive usage and observed request/response URLs
PASS strict mismatch is visible with original GPT identifier and one request, no retry or tool dispatch
PASS 429 preserves native type, raw message, request ID, URL and inputs without automatic replay
{"success":true,"checks":7,"directory":"D:\\Ai-project\\REA逆向\\Agent-Workbench\\.qa\\messages-ui-m-fae17b5c-0946-44b2-b655-f17521ae9af4"}
@@ -0,0 +1,26 @@
✔ full candidate validation returns the exact loaded-input host identity and rejects undeclared or altered files (756.3716ms)
✔ installation requires separate absolute program/data directories and a candidate in its own versions namespace (897.9194ms)
✔ staging retains the original data and pointer without querying or restarting a host (2836.904ms)
✔ a post-pointer failure restores every original pointer byte and persistent row; a first-install failure removes only its published pointer (9940.0255ms)
✔ changed active manifest and concurrent pointer changes are rejected without overwriting the unknown record (3549.5156ms)
✔ an unregistered non-HTTP listener blocks preflight even when service-info is absent (2226.3433ms)
✔ uninstall archives its original pointer and preserves every data row/artifact instead of recursively deleting data (5724.6649ms)
✔ preflight rejects an unrelated program root or changed native bootstrap while preserving its original bytes (3848.1167ms)
✔ a changed owned bootstrap follows its candidate and a post-pointer fault restores both original bytes (14472.7208ms)
✔ an external bootstrap change after preflight is preserved and never overwritten as an update or rollback (6816.9042ms)
✔ a live SQLite host with the actual 1.9 artifact object upgrades, keeps its original blobs and verifies real browser PNG bytes (3573.7497ms)
✔ failed upgrade restores the original artifact object, provider and request handler (1391.4458ms)
✔ hot upgrade adopts a retained virtual data root without reopening SQLite or changing rows, Provider or routing (1160.4632ms)
✔ failed adoption restores the original virtual data-root name and original open database (523.9064ms)
✔ hot upgrade refreshes retained model and importer methods while preserving credential and routing objects (596.7795ms)
✔ hot upgrade refuses active probes and restores retained model methods if installation fails (482.7736ms)
✔ a directly started compiled host upgrades twice with the same PID, sessions, applied bytes and volatile provider (3419.6486ms)
✔ the running 0.1.0 process upgrades in place and retains its project, session, files and volatile model connection (6547.8833ms)
ℹ tests 18
ℹ suites 0
ℹ pass 18
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 69547.9324
@@ -0,0 +1,71 @@
import fs from 'node:fs';
import path from 'node:path';
import assert from 'node:assert/strict';
import { DatabaseSync } from 'node:sqlite';
import { digest, hash, canonical } from './desktop-f-state.mjs';
const root = path.resolve('.qa/messages-m/live'), read = n => JSON.parse(fs.readFileSync(path.join(root, n)));
const before = read('before.json'), after = read('after.json'), setup = read('setup.json'), owned = read('owned.json'), attempts = read('attempts.json'), result = read('result.json');
assert.equal(after.pid, before.pid); assert.equal(after.hasKey, true); assert.deepEqual(after.active, []);
assert.equal(after.configSha256, before.configSha256); assert.deepEqual(after.modelCenter.selected, before.modelCenter.selected);
assert.deepEqual(after.exports.filter(x => x.id !== owned.session.id), before.exports, 'Every original full export exactly retained.');
assert.equal(after.projects, before.projects + 1); assert.equal(after.database.sessionIds.length, before.database.sessionIds.length + 1);
for (const file of before.persistent.files) assert.deepEqual(after.persistent.files.find(f => f.path === file.path), file, 'Original persistent file ' + file.path);
for (const file of before.persistent.artifacts) assert.deepEqual(after.persistent.artifacts.find(f => f.path === file.path), file, 'Original artifact ' + file.path);
const db = new DatabaseSync(path.join(setup.dataRoot, 'workbench.sqlite'), { readOnly: true });
const checks = [], additions = [];
try {
db.exec('BEGIN');
for (const [name, old] of Object.entries(before.database.tables)) {
assert.equal(after.database.tables[name].schemaSha256, old.schemaSha256, name + ' schema');
const current = db.prepare('SELECT * FROM "' + name + '" ORDER BY rowid').all().filter(r => !(name === 'metadata' && r.key === 'runtimeOwner'));
const key = (r, index) => r.id ?? r.key ?? r.seq ?? index, oldIds = new Set(old.rows.map(r => r.id));
let exact = 0, declared = 0;
for (const previous of old.rows) {
const row = name === 'sqlite_sequence' ? current.find(r => r.name === 'events') : current.find((r, i) => key(r, i) === previous.id);
assert.ok(row, name + '/' + previous.id + ' remains');
if (digest(row) === previous.sha256) { exact++; continue; }
let normalized;
if (name === 'projects') normalized = { ...row, position: row.position - 1 };
else if (name === 'metadata' && row.key === 'modelRegistry') {
const actual = JSON.parse(row.value), original = setup.metadata.modelRegistry, previousConnection = original.connections.find(c => c.id === setup.originalConnection.id);
const connection = actual.connections.find(c => c.id === previousConnection.id);
assert.equal(actual.revision, original.revision + 1); assert.equal(connection.revision, previousConnection.revision + 1);
assert.equal(connection.protocol, 'anthropic-messages'); assert.equal(connection.messagesAuth, 'bearer');
assert.equal(connection.modelIdentityPolicy, 'require-match'); assert.deepEqual(connection.modelIdentityAliases, []); assert.equal(connection.modelIdentityAliasSource, '');
assert.deepEqual(connection.nativeParameters, {}); assert.equal(connection.outputTokenField, 'max_tokens'); assert.ok(connection.updatedAt > previousConnection.updatedAt);
const target = structuredClone(connection);
for (const field of ['protocol', 'messagesAuth', 'modelIdentityPolicy', 'modelIdentityAliases', 'modelIdentityAliasSource', 'nativeParameters', 'outputTokenField', 'revision', 'updatedAt']) {
if (Object.hasOwn(previousConnection, field)) target[field] = previousConnection[field]; else delete target[field];
}
assert.deepEqual(canonical(target), canonical(previousConnection), 'Only declared target connection fields changed.');
actual.connections[actual.connections.findIndex(c => c.id === previousConnection.id)] = previousConnection; actual.revision = original.revision;
assert.deepEqual(canonical(actual), canonical(original)); normalized = { key: row.key, value: JSON.stringify(original) };
} else if (name === 'metadata' && row.key === 'modelAttempts') {
const actual = JSON.parse(row.value), added = actual.filter(a => a.sessionId === owned.session.id);
assert.deepEqual(added, attempts); assert.equal(added.length, 1);
assert.deepEqual(actual.filter(a => a.sessionId !== owned.session.id), setup.metadata.modelAttempts);
normalized = { key: row.key, value: JSON.stringify(setup.metadata.modelAttempts) };
} else if (name === 'sqlite_sequence' && row.name === 'events') {
assert.equal(row.seq, after.eventCursor); normalized = { ...row, seq: before.eventCursor };
} else assert.fail('Unexplained original row mutation ' + name + '/' + previous.id);
assert.equal(digest(normalized), previous.sha256, name + '/' + previous.id + ' original SHA reconstructs exactly'); declared++;
}
for (const [index, row] of current.entries()) if (name !== 'sqlite_sequence' && !oldIds.has(key(row, index))) {
if (name === 'projects') assert.equal(row.id, owned.project.id);
else if (name === 'sessions') assert.equal(row.id, owned.session.id);
else if (['records', 'events'].includes(name)) assert.equal(row.session_id ?? row.sessionId, owned.session.id);
else if (name === 'outbox') assert.equal(db.prepare('SELECT session_id FROM events WHERE seq=?').get(row.seq)?.session_id, owned.session.id);
else assert.fail('Unexplained new row ' + name);
additions.push({ table: name, id: key(row, index), sha256: digest(row) });
}
checks.push({ table: name, original: old.count, current: current.length, exact, declared });
}
db.exec('ROLLBACK');
} finally { db.close(); }
assert.equal(result.requests, 1); assert.equal(result.actualTools, 0); assert.equal(result.toolDefinitions, 0); assert.equal(result.credentialRevisionRetained, true);
const receipt = { passed: true, baselineSha256: hash(fs.readFileSync(path.join(root, 'before.json'))), afterSha256: hash(fs.readFileSync(path.join(root, 'after.json'))),
checks, additions, originalExportsExact: before.exports.length, originalArtifactBytesExact: before.persistent.artifacts.length,
originalPersistentFilesExact: before.persistent.files.length, globalConfigAndSelectionExact: true, noCredentialExtractionOrResave: true, originalPidRetained: true,
declaredChanges: ['One owned project/session; original project positions shift exactly by one.', 'One declared connection protocol/auth/native-default/strict-identity revision; all other registry and connection fields exact.', 'One owned attempt appended; original attempt records exact.', 'Event sequence increases only for the owned session.'] };
fs.writeFileSync(path.join(root, 'state-audit.json'), JSON.stringify(receipt, null, 2) + '\n', { flag: 'wx' });
console.log(JSON.stringify({ passed: true, checks, originalExports: receipt.originalExportsExact, originalArtifacts: receipt.originalArtifactBytesExact, newRows: additions.length }));
@@ -0,0 +1,9 @@
import fs from 'node:fs';
import assert from 'node:assert/strict';
const read = p => JSON.parse(fs.readFileSync(p));
const original = read('.qa/messages-m/product-freeze-release.json'), final = read('.qa/messages-m/product-freeze-final.json');
const files = original.files.filter(f => f.path.startsWith('src/')).map(f => { const current = final.files.find(c => c.path === f.path); assert.deepEqual(current, f, f.path); return f; });
const result = { passed: true, fixtureSourceCommit: original.sourceCommit, finalSourceCommit: final.sourceCommit, files,
scope: 'All product src/ input bytes are identical between the tested probe driver build and final build; only the retained-host upgrade version whitelist and documentation changed. The real preflight still executes the final installed original Provider.' };
fs.writeFileSync('.qa/messages-m/probe-source-binding.json', JSON.stringify(result, null, 2), { flag: 'wx' });
console.log(JSON.stringify({ passed: true, identicalProductSourceFiles: files.length }));
@@ -0,0 +1,15 @@
import fs from 'node:fs';
import path from 'node:path';
import assert from 'node:assert/strict';
import { createHash } from 'node:crypto';
import { execFileSync } from 'node:child_process';
const root = path.resolve('.qa/messages-m'); fs.mkdirSync(root, { recursive: true });
assert.equal(execFileSync('git', ['status', '--porcelain'], { encoding: 'utf8' }).trim(), '');
const names = execFileSync('git', ['ls-files', '-z', '--', 'src', 'public', 'scripts', 'tests', 'examples', 'benchmarks', 'research', 'package.json', 'package-lock.json', 'vite.config.ts', 'tsconfig.json', 'index.html', 'mobile.html'], { encoding: 'utf8' }).split('\0').filter(Boolean).filter(name =>
/^(?:src|public|scripts|tests|examples|benchmarks|research)\//.test(name) || ['package.json','package-lock.json','vite.config.ts','tsconfig.json','index.html','mobile.html'].includes(name));
const record = file => { const bytes = fs.readFileSync(file); return { path: file.replaceAll('\\', '/'), bytes: bytes.length, sha256: createHash('sha256').update(bytes).digest('hex') }; };
const walk = folder => fs.readdirSync(folder, { withFileTypes: true }).flatMap(entry => entry.isDirectory() ? walk(path.join(folder, entry.name)) : [path.join(folder, entry.name)]);
const result = { sourceCommit: execFileSync('git', ['rev-parse', 'HEAD'], { encoding: 'utf8' }).trim(), version: '1.24.4', createdAt: new Date().toISOString(),
files: names.sort().map(record), compiled: ['dist-client','dist-server','dist-desktop'].flatMap(walk).sort().map(record), realModelRequests: 0 };
fs.writeFileSync(path.join(root, 'product-freeze-final.json'), JSON.stringify(result, null, 2) + '\n', { flag: 'wx' });
console.log(JSON.stringify({ success: true, files: result.files.length, compiled: result.compiled.length, sourceCommit: result.sourceCommit }));

Some files were not shown because too many files have changed in this diff Show More