test(ai-coach): 轻量架构可行性评测(方案 §1–17)+ 第一轮结果 CONDITIONAL GO - #546
Open
sol-wizard wants to merge 1 commit into
Open
sol-wizard wants to merge 1 commit into
sol-wizard wants to merge 1 commit into
Conversation
…plan §1–17) Layer 1: 12 scripted architecture cases (A1–A12) plus variants and 4 known-gap characterizations, run through the real ConversationKernel and ModelTurnRuntime with a scripted gateway. Layer 2: 16 live-model cases (E1–E8, C1–C8) with hard checks, a lightweight judge and an AICOACH_EVAL_RUNS stability pass; 12-sample judge calibration. One JSON record per case run under TestResults/ai-coach-evals/. Round-1 verdict: CONDITIONAL GO (see blotz-v2 round-1 report). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01YMgJV8JoQ93ZXL8kEuuzry
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
按《AI Action Coach 轻量架构可行性评测方案》§1–§17 实现的评测代码,以及第一轮运行结果。§18(failure bundle / 自动修复闭环)没做。
第一轮结论:CONDITIONAL GO
授权边界全部守住(0 张未授权卡片、0 条虚构 Evidence 获得授权、0 次未确认的正式变更、提问规则 0 违反),多轮状态由 Kernel 正确表达。但有两处「守得太死、不给纠错机会」要先修:
_KnownGapcharacterization必须先修
A3b)。E1「帮我安排明天下午三点整理论文资料,做半小时」模型卡片完全正确,只因 constraint 文本
明天下午三点整比引用明天下午三点多一个字,HasInvalidClaims→ConversationPostPolicy第 55–62 行Downgrade(ContinueListening, EvidenceInvalid)→ 用户看到「抱歉,这次没能形成合适的回复」。249d2cca(09-12「optimize」)把这里从RequiresRegeneration改成了Downgrade;PostPolicy_InvalidCurrentClaimWithActiveIntent_RejectsModelProposal从那天起就是红的。建议:constraint claim 不可验证时只丢该条 constraint(item / actionRequest / disposition 的虚构仍致命),并/或恢复 RequiresRegeneration。
direct_instruction(C8 三次里一次)。模型把「…但还是帮我安排明天早上跑步吧」标成explicit_planning_request,ProposalTrigger = CurrentTurnDirectInstruction就拒绝出卡。建议触发同时接受explicit_planning_request(仍要求引用来自当前消息)。其他发现(未声明的问句绕过提问节奏、有约束的 intent 拿不到确定性兜底、部分纠正与整体拒绝不可区分、降级时模型文本整段换成 canned 道歉等)见完整报告
blotz-v2/ai-action-coach-evaluation-round1-report.md。这个 PR 里有什么
全部在
blotztask-test/AiCoach/Evaluation/,不动生产代码:AiCoachEvalHarness:真ConversationKernel+ 真ModelTurnRuntime(prompt、context builder、Evidence Guard、Planning/Support Policy、Post-Policy、Guards),只换 gateway;固定时间 2026-09-14 周一 10:00 Sydney。AiCoachEvalCase/EvalChecks/QuestionQualityEvaluator:Required / Allowed / Forbidden 契约与硬规则检查,断言公开 Outcome 和 Kernel 状态,不解析日志。LightweightJudge+QuestionJudgeCalibrationTests:同一部署做 judge,只出软信号,12 样本校准。EvalRunRecorder:每个 Case 每次运行一行 JSON 到TestResults/ai-coach-evals/(已 gitignore),含原始模型输出。ArchitectureFeasibilityTests(A1–A12 + 变体 + 已知缺口)、ExecutionModeEvalTests(E1–E8)、CompanionModeEvalTests(C1–C8)。真模型层需要
blotztask-api/appsettings.Development.json的 Azure OpenAI 凭据,没有则显示 SKIPPED、不算通过。注意
ai-coach-v3-pro上原本就有 7 个红测试(ModelTurnRuntimeTests6 个 +ConversationPolicyTests1 个),本 PR 没有改它们:6 个是旧 scripted 候选缺actionRequest字段、被2f288aad后的 readiness 规则降级;1 个就是上面阻塞 1 的规约冲突。_KnownGap测试固定的是当前行为;修好后会翻红,届时改成正向断言。🤖 Generated with Claude Code
https://claude.ai/code/session_01YMgJV8JoQ93ZXL8kEuuzry