fix: expose workbench timing source status

This commit is contained in:
Codex
2026-06-28 11:24:36 +00:00
parent f4d97774cf
commit c80e85f4b4
6 changed files with 172 additions and 31 deletions
@@ -259,6 +259,8 @@ console.log(JSON.stringify({
sampleDeltaSeconds: item.sampleDeltaSeconds ?? null,
allowedIncreaseSeconds: item.allowedIncreaseSeconds ?? null,
excessiveIncreaseSeconds: item.excessiveIncreaseSeconds ?? null,
timingSourceOfTruth: item.timingSourceOfTruth ?? null,
timingStatus: item.timingStatus ?? null,
})),
turnTimingElapsedZeroResets: recentWindow.sampleMetrics.turnTimingElapsedZeroResets.slice(0, 8).map((item) => ({
columnLabel: item.columnLabel ?? item.columnId ?? null,
@@ -275,6 +277,8 @@ console.log(JSON.stringify({
toValue: item.toValue ?? null,
delta: item.delta ?? null,
anomaly: item.anomaly ?? null,
timingSourceOfTruth: item.timingSourceOfTruth ?? null,
timingStatus: item.timingStatus ?? null,
})),
turnTimingTotalElapsedForwardJumps: recentWindow.sampleMetrics.turnTimingTotalElapsedForwardJumps.slice(0, 8).map((item) => ({
columnLabel: item.columnLabel ?? item.columnId ?? null,
@@ -294,12 +298,14 @@ console.log(JSON.stringify({
allowedIncreaseSeconds: item.allowedIncreaseSeconds ?? null,
excessiveIncreaseSeconds: item.excessiveIncreaseSeconds ?? null,
anomaly: item.anomaly ?? null,
timingSourceOfTruth: item.timingSourceOfTruth ?? null,
timingStatus: item.timingStatus ?? null,
})),
pagePerformanceSlowApi: recentWindow.pagePerformance.sameOriginApiByPath.filter((item) => item.isLongLivedStream !== true && Number(item.overBudgetCount ?? item.overFiveSecondCount ?? 0) > 0).slice(0, 5).map((item) => ({ path: item.path, route: item.route, sampleCount: item.sampleCount, p95Ms: item.p95Ms, maxMs: item.maxMs, overBudgetCount: item.overBudgetCount, budgetMs: item.budgetMs, overFiveSecondCount: item.overFiveSecondCount, slowSamples: item.slowSamples })),
archivePagePerformanceSlowApi: pagePerformance.sameOriginApiByPath.filter((item) => item.isLongLivedStream !== true && Number(item.overBudgetCount ?? item.overFiveSecondCount ?? 0) > 0).slice(0, 8).map((item) => ({ path: item.path, route: item.route, sampleCount: item.sampleCount, p95Ms: item.p95Ms, maxMs: item.maxMs, overBudgetCount: item.overBudgetCount, budgetMs: item.budgetMs, overFiveSecondCount: item.overFiveSecondCount, slowSamples: item.slowSamples })),
pagePerformancePartialApi: recentWindow.pagePerformance.sameOriginApiByPath.filter((item) => item.isLongLivedStream !== true && Number(item.partialOverBudgetCount ?? item.partialOverFiveSecondCount ?? 0) > 0).slice(0, 5).map((item) => ({ path: item.path, route: item.route, sampleCount: item.sampleCount, completeTimingSampleCount: item.completeTimingSampleCount, partialTimingSampleCount: item.partialTimingSampleCount, partialOverBudgetCount: item.partialOverBudgetCount, budgetMs: item.partialBudgetMs, partialOverFiveSecondCount: item.partialOverFiveSecondCount, partialSamples: item.partialSamples })),
pagePerformanceSseStreams: recentWindow.pagePerformance.sameOriginApiByPath.filter((item) => item.isLongLivedStream === true).slice(0, 5).map((item) => ({ path: item.path, route: item.route, sampleCount: item.sampleCount, streamOpenSampleCount: item.streamOpenSampleCount, streamOpenP95Ms: item.streamOpenP95Ms, streamOpenMaxMs: item.streamOpenMaxMs, streamOpenOverBudgetCount: item.streamOpenOverBudgetCount, streamOpenBudgetMs: item.streamOpenBudgetMs, streamOpenOverFiveSecondCount: item.streamOpenOverFiveSecondCount, streamLifetimeOverFiveSecondCount: item.streamLifetimeOverFiveSecondCount, slowSamples: item.slowSamples })),
findings: prioritizeFindings(recentWindow.findings).slice(0, 8).map((item) => ({ kind: item.id ?? item.kind ?? item.code, severity: item.severity, count: item.count ?? item.sampleCount ?? null, summary: String(item.summary ?? item.message ?? "").slice(0, 180) })),
findings: prioritizeFindings(recentWindow.findings).slice(0, 8).map((item) => ({ kind: item.id ?? item.kind ?? item.code, severity: item.severity, count: item.count ?? item.sampleCount ?? null, timingSourceOfTruth: item.timingSourceOfTruth ?? null, timingStatus: item.timingStatus ?? null, summary: String(item.summary ?? item.message ?? "").slice(0, 180) })),
traceOrderAnomalies: (recentWindow.sampleMetrics?.traceOrder?.orderAnomalies || []).slice(0, 8).map((item) => ({
sampleSeq: item.sampleSeq ?? null,
sampleIndex: item.sampleIndex ?? null,
@@ -345,6 +351,8 @@ console.log(JSON.stringify({
cardTotalElapsedSeconds: item.cardTotalElapsedSeconds ?? null,
deltaSeconds: item.deltaSeconds ?? null,
toleranceSeconds: item.toleranceSeconds ?? null,
timingSourceOfTruth: item.timingSourceOfTruth ?? null,
timingStatus: item.timingStatus ?? null,
})),
codeAgentCardDurationUnderreported: (recentWindow.sampleMetrics?.codeAgentCardTiming?.durationUnderreported || []).slice(0, 8).map((item) => ({
sampleIndex: item.sampleIndex ?? null,
@@ -355,6 +363,8 @@ console.log(JSON.stringify({
expectedElapsedSeconds: item.expectedElapsedSeconds ?? null,
deltaSeconds: item.deltaSeconds ?? null,
evidenceKind: item.evidenceKind ?? null,
timingSourceOfTruth: item.timingSourceOfTruth ?? null,
timingStatus: item.timingStatus ?? null,
evidencePreview: String(item.evidencePreview || "").slice(0, 120),
})),
codeAgentCardDurationMismatches: (recentWindow.sampleMetrics?.codeAgentCardTiming?.durationMismatches || []).slice(0, 8).map((item) => ({
@@ -369,6 +379,8 @@ console.log(JSON.stringify({
deltaSeconds: item.deltaSeconds ?? null,
evidenceKind: item.evidenceKind ?? null,
exactEvidence: item.exactEvidence === true,
timingSourceOfTruth: item.timingSourceOfTruth ?? null,
timingStatus: item.timingStatus ?? null,
evidencePreview: String(item.evidencePreview || "").slice(0, 120),
})),
valuesRedacted: true,
@@ -2747,6 +2759,9 @@ function buildFindings(samples, control, network, errors, sampleMetrics, promptN
id: "code-agent-card-duration-underreported",
severity: "amber",
summary: "completed Code Agent card total elapsed is shorter than trace/final-response duration evidence; timing mismatch is a non-blocking alert",
timingSourceOfTruth: "trace-completion-total-or-final-response-duration",
timingStatus: timingStatusFromRows(cardTiming.durationUnderreported, "business-turn-completed"),
timingAlert: true,
count: cardTimingSummary.durationUnderreportedCount,
samples: (cardTiming.durationUnderreported || []).slice(0, 20),
});
@@ -2756,6 +2771,9 @@ function buildFindings(samples, control, network, errors, sampleMetrics, promptN
id: "code-agent-card-duration-mismatch",
severity: "amber",
summary: "completed Code Agent card total elapsed does not match sealed completion/final-response timing evidence; timing mismatch is a non-blocking alert",
timingSourceOfTruth: "trace-completion-total-or-final-response-duration",
timingStatus: timingStatusFromRows(cardTiming.durationMismatches, "business-turn-completed"),
timingAlert: true,
count: cardTimingSummary.durationMismatchCount,
samples: (cardTiming.durationMismatches || []).slice(0, 20),
});
@@ -2780,7 +2798,7 @@ function buildFindings(samples, control, network, errors, sampleMetrics, promptN
samples: (traceOrder.completionNotLast || []).slice(0, 20),
});
}
if (Number(cardTimingSummary.roundCompletionElapsedMismatchCount ?? 0) > 0) findings.push({ id: "round-completion-elapsed-mismatch", severity: "amber", summary: "Trace row 轮次完成(总耗时 ... does not match the visible Code Agent card total elapsed time within YAML timing slack; timing mismatch is a non-blocking alert", count: cardTimingSummary.roundCompletionElapsedMismatchCount, toleranceSeconds: cardTimingSummary.elapsedMismatchToleranceSeconds, samples: (roundCompletion.elapsedMismatches || []).slice(0, 20) });
if (Number(cardTimingSummary.roundCompletionElapsedMismatchCount ?? 0) > 0) findings.push({ id: "round-completion-elapsed-mismatch", severity: "amber", summary: "Trace row 轮次完成(总耗时 ... does not match the visible Code Agent card total elapsed time within YAML timing slack; timing mismatch is a non-blocking alert", timingSourceOfTruth: "trace-round-completion-total", timingStatus: timingStatusFromRows(roundCompletion.elapsedMismatches, "business-turn-completed"), timingAlert: true, count: cardTimingSummary.roundCompletionElapsedMismatchCount, toleranceSeconds: cardTimingSummary.elapsedMismatchToleranceSeconds, samples: (roundCompletion.elapsedMismatches || []).slice(0, 20) });
if (Number(cardTimingSummary.roundCompletionFinalResponseMissingCount ?? 0) > 0) findings.push({ id: "round-completion-final-response-missing", severity: "red", summary: "Trace row showed 轮次完成, but no final response was visible in the Code Agent card afterward", count: cardTimingSummary.roundCompletionFinalResponseMissingCount, samples: (roundCompletion.finalResponseMissing || []).slice(0, 20) });
if (Number(cardTimingSummary.roundCompletionPostTimingChangeCount ?? 0) > 0) findings.push({ id: "round-completion-post-timing-change", severity: "amber", summary: "After 轮次完成, card total elapsed or 最近更新 continued changing; terminal timing alert is non-blocking", count: cardTimingSummary.roundCompletionPostTimingChangeCount, samples: (roundCompletion.postCompletionTimingChanges || []).slice(0, 20) });
if (Number(cardTimingSummary.roundCompletionPostRecentUpdateVisibleCount ?? 0) > 0) findings.push({ id: "round-completion-recent-update-still-visible", severity: "info", summary: "最近更新 was still visible after 轮次完成; inspect whether terminal cards should hide activity age or keep it sealed", count: cardTimingSummary.roundCompletionPostRecentUpdateVisibleCount, samples: (roundCompletion.postCompletionRecentUpdateVisible || []).slice(0, 20) });
@@ -2824,15 +2842,15 @@ function buildFindings(samples, control, network, errors, sampleMetrics, promptN
const promptSteerRounds = Array.isArray(promptNetwork?.rounds) ? promptNetwork.rounds.filter((item) => item.steerUsed === true) : [];
if (promptSteerRounds.length > 0) findings.push({ id: "prompt-routed-to-steer", severity: "amber", summary: "sendPrompt was submitted through /v1/agent/chat/steer; verify the previous turn was truly in-flight and not an unsealed terminal failure", count: promptSteerRounds.length, rounds: promptSteerRounds.slice(0, 10) });
const elapsedZeroResets = Array.isArray(sampleMetrics?.turnTimingElapsedZeroResets) ? sampleMetrics.turnTimingElapsedZeroResets : [];
if (elapsedZeroResets.length > 0) findings.push({ id: "turn-timing-total-elapsed-zero-reset", severity: "amber", summary: "Code Agent total elapsed jumped from a non-zero value back to 0 seconds; timing reset is a non-blocking alert", count: elapsedZeroResets.length, samples: elapsedZeroResets.slice(0, 20) });
if (elapsedZeroResets.length > 0) findings.push({ id: "turn-timing-total-elapsed-zero-reset", severity: "amber", summary: "Code Agent total elapsed jumped from a non-zero value back to 0 seconds; timing reset is a non-blocking alert", timingSourceOfTruth: "dom-card-total-elapsed-sequence", timingStatus: timingStatusFromRows(elapsedZeroResets), timingAlert: true, count: elapsedZeroResets.length, samples: elapsedZeroResets.slice(0, 20) });
const elapsedDecreases = Array.isArray(sampleMetrics?.turnTimingNonMonotonic)
? sampleMetrics.turnTimingNonMonotonic.filter((item) => item.metric === "totalElapsedSeconds" && item.anomaly !== "zero-reset")
: [];
if (elapsedDecreases.length > 0) findings.push({ id: "turn-timing-total-elapsed-decrease", severity: "amber", summary: "Code Agent total elapsed decreased between adjacent samples; timing decrease is a non-blocking alert", count: elapsedDecreases.length, samples: elapsedDecreases.slice(0, 20) });
if (elapsedDecreases.length > 0) findings.push({ id: "turn-timing-total-elapsed-decrease", severity: "amber", summary: "Code Agent total elapsed decreased between adjacent samples; timing decrease is a non-blocking alert", timingSourceOfTruth: "dom-card-total-elapsed-sequence", timingStatus: timingStatusFromRows(elapsedDecreases), timingAlert: true, count: elapsedDecreases.length, samples: elapsedDecreases.slice(0, 20) });
const elapsedForwardJumps = Array.isArray(sampleMetrics?.turnTimingTotalElapsedForwardJumps) ? sampleMetrics.turnTimingTotalElapsedForwardJumps : [];
if (elapsedForwardJumps.length > 0) findings.push({ id: "turn-timing-total-elapsed-forward-jump", severity: "amber", summary: "Code Agent total elapsed jumped forward faster than browser sample interval; timing jump is a non-blocking alert", count: elapsedForwardJumps.length, samples: elapsedForwardJumps.slice(0, 20) });
if (elapsedForwardJumps.length > 0) findings.push({ id: "turn-timing-total-elapsed-forward-jump", severity: "amber", summary: "Code Agent total elapsed jumped forward faster than browser sample interval; timing jump is a non-blocking alert", timingSourceOfTruth: "dom-card-total-elapsed-sequence", timingStatus: timingStatusFromRows(elapsedForwardJumps), timingAlert: true, count: elapsedForwardJumps.length, samples: elapsedForwardJumps.slice(0, 20) });
const terminalElapsedGrowth = Array.isArray(sampleMetrics?.turnTimingTerminalElapsedGrowth) ? sampleMetrics.turnTimingTerminalElapsedGrowth : [];
if (terminalElapsedGrowth.length > 0) findings.push({ id: "turn-timing-terminal-elapsed-growth", severity: "amber", summary: "terminal Code Agent card total elapsed changed after terminal status; terminal timing alert is non-blocking", count: terminalElapsedGrowth.length, samples: terminalElapsedGrowth.slice(0, 20) });
if (terminalElapsedGrowth.length > 0) findings.push({ id: "turn-timing-terminal-elapsed-growth", severity: "amber", summary: "terminal Code Agent card total elapsed changed after terminal status; terminal timing alert is non-blocking", timingSourceOfTruth: "terminal-card-total-elapsed-seal", timingStatus: timingStatusFromRows(terminalElapsedGrowth, "business-turn-completed"), timingAlert: true, count: terminalElapsedGrowth.length, samples: terminalElapsedGrowth.slice(0, 20) });
const recentUpdateSawtoothJumps = Array.isArray(sampleMetrics?.turnTimingRecentUpdateSawtoothJumps)
? sampleMetrics.turnTimingRecentUpdateSawtoothJumps
: Array.isArray(sampleMetrics?.turnTimingNonMonotonic)
@@ -2841,7 +2859,7 @@ function buildFindings(samples, control, network, errors, sampleMetrics, promptN
if (recentUpdateSawtoothJumps.length > 0) findings.push({ id: "turn-timing-recent-update-sawtooth-jump", severity: "amber", summary: "最近更新 value jumped faster than sample interval; expected sawtooth increase-or-reset", count: recentUpdateSawtoothJumps.length, samples: recentUpdateSawtoothJumps.slice(0, 20) });
const severeTimeoutRounds = Array.isArray(sampleMetrics?.rounds) ? sampleMetrics.rounds.filter((item) => Number(item.maxTotalElapsedSeconds) > alertThresholds.turnElapsedSevereTimeoutSeconds) : [];
const severeTimeoutSamples = Array.isArray(sampleMetrics?.timeline) ? sampleMetrics.timeline.filter((item) => Number(item.totalElapsedSeconds) > alertThresholds.turnElapsedSevereTimeoutSeconds) : [];
if (severeTimeoutRounds.length > 0 || severeTimeoutSamples.length > 0) findings.push({ id: "turn-elapsed-severe-timeout", severity: "amber", summary: "turn total elapsed exceeded the YAML-configured elapsed alert threshold; timing is a non-blocking alert unless the turn fails to complete or breaks multi-round continuity", thresholdSeconds: alertThresholds.turnElapsedSevereTimeoutSeconds, count: Math.max(severeTimeoutRounds.length, severeTimeoutSamples.length), rounds: severeTimeoutRounds.slice(0, 20), samples: severeTimeoutSamples.slice(0, 20) });
if (severeTimeoutRounds.length > 0 || severeTimeoutSamples.length > 0) findings.push({ id: "turn-elapsed-severe-timeout", severity: "amber", summary: "turn total elapsed exceeded the YAML-configured elapsed alert threshold; timing is a non-blocking alert unless the turn fails to complete or breaks multi-round continuity", timingSourceOfTruth: "dom-card-total-elapsed-yaml-threshold", timingStatus: timingStatusFromRows([...severeTimeoutRounds, ...severeTimeoutSamples], "observer-timeout"), timingAlert: true, thresholdSeconds: alertThresholds.turnElapsedSevereTimeoutSeconds, count: Math.max(severeTimeoutRounds.length, severeTimeoutSamples.length), rounds: severeTimeoutRounds.slice(0, 20), samples: severeTimeoutSamples.slice(0, 20) });
const loadingSummary = sampleMetrics?.loading?.summary || {};
const visibleLoadingSlowSeconds = alertThresholds.visibleLoadingSlowMs / 1000;
if (Number(loadingSummary.longestContinuousSeconds ?? 0) > visibleLoadingSlowSeconds) findings.push({ id: "page-loading-visible-over-budget", severity: "red", summary: "visible 加载中 stayed on screen longer than configured YAML budget; fix real loading latency instead of revealing incomplete content early", count: loadingSummary.overBudgetSegmentCount ?? loadingSummary.overFiveSecondSegmentCount ?? 1, longestContinuousSeconds: loadingSummary.longestContinuousSeconds, budgetSeconds: visibleLoadingSlowSeconds, segments: sampleMetrics.loading.segments.slice(0, 20), owners: sampleMetrics.loading.owners.slice(0, 20) });