robbond
956fc2e31e
Two bugs were causing the model to return {"status":"ok"} / {"status":"ready"}
instead of structured reconstruction data, resulting in POST /api/analyse 500:
1. DOUBLE-WRAPPING BUG (lib/llm/provider.js):
generateReconstruction() called buildPrompt(scenario) on input that was
already a fully-built prompt string from analyseScenario(). This wrapped the
v0.1 prompt (~5000+ chars) in another template layer, producing incomprehensible
output that the model could not parse as structured JSON.
Fix: Pass scenario through directly (it is ALREADY a built prompt).
2. MISSING JSON SPEC (prompts/reconstruct-v0.2.md):
The v0.2 prompt template said 'matching the structure exactly' but never
defined what that structure was. The model invented its own field names
(input_classification, reasoning_mode, anchors) with snake_case instead of
camelCase, which failed Zod validation -> 500 errors.
Fix: Added explicit JSON schema section with exact key names, enum values,
and nested structure matching the Zod validation layer.
Additionally:
- Refactored route to use analyseScenario from lib/analysis (centralized)
- Added lib/analysis.js with shared analysis logic
- Updated components to display promptVersion and validation errors
- Added lib/reconstruction/prompt.js v0.1/v0.2 versioning
- Added lib/reconstruction/schema.js v0.2 Zod schemas
- Added debug tool scripts, evaluation results, and comparison findings
585 lines
15 KiB
JSON
585 lines
15 KiB
JSON
{
|
|
"timestamp": "2026-08-01T06:01:56.070Z",
|
|
"provider": "mock",
|
|
"promptVersion": "v0.2",
|
|
"casesRun": 10,
|
|
"summary": {
|
|
"technical": {
|
|
"schemaValidityRate": "100.0%",
|
|
"classificationMatchRate": "80.0%",
|
|
"nextQuestionPresentRate": "100.0%",
|
|
"passRate": "80.0%"
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConceptMatchRate": "0.0%",
|
|
"unsupportedInferenceFailures": "0",
|
|
"passRate": "0.0%"
|
|
},
|
|
"combinedPassRate": "0.0%",
|
|
"averageResponseDurationMs": "2"
|
|
},
|
|
"testCaseResults": [
|
|
{
|
|
"id": "diag-01",
|
|
"input": "We've seen a spike in complaints from our warehouse team this month compared to last month.",
|
|
"responseDurationMs": 6,
|
|
"actualPrimaryType": "unexplained_change",
|
|
"actualReasoningModes": [
|
|
"establish_baseline",
|
|
"identify_difference"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": true,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": true,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "complaints",
|
|
"found": true
|
|
},
|
|
{
|
|
"concept": "warehouse",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "baseline comparison",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "quality issue",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "staff turnover",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "training gap",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-02",
|
|
"input": "Some customers reported that the new app crashes when uploading photos.",
|
|
"responseDurationMs": 2,
|
|
"actualPrimaryType": "observed_problem",
|
|
"actualReasoningModes": [
|
|
"identify_difference"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": true,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": true,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "app crashes",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "photo upload",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "some customers",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "all users affected",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "server-side bug",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "Android only",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-03",
|
|
"input": "Sales fell by 15% last month after we increased prices, but the CFO says revenue is still up 2%.",
|
|
"responseDurationMs": 1,
|
|
"actualPrimaryType": "causal_claim",
|
|
"actualReasoningModes": [
|
|
"establish_baseline",
|
|
"identify_difference"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": false,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": false,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "sales decline",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "price increase",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "revenue increase",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "CFO report",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "price was set too high",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "competitors gained market share",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "revenue data is wrong",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-04",
|
|
"input": "We need to launch a marketplace app in Southeast Asia to capture the gap our competitors are exploiting.",
|
|
"responseDurationMs": 0,
|
|
"actualPrimaryType": "decision_request",
|
|
"actualReasoningModes": [
|
|
"identify_difference",
|
|
"decision_support",
|
|
"identify_missing_information"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": true,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": true,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "marketplace app",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "Southeast Asia",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "competitor gap",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "this will definitely succeed",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "we have the resources",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "competitors are struggling",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-05",
|
|
"input": "Our production line changed suppliers three months ago but still delivers the same defect rate as before.",
|
|
"responseDurationMs": 1,
|
|
"actualPrimaryType": "unexplained_change",
|
|
"actualReasoningModes": [
|
|
"identify_difference",
|
|
"establish_baseline",
|
|
"validate_measurement"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": true,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": true,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "supplier change",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "three months ago",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "same defect rate",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "new supplier is worse",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "old supplier was better",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "quality process is broken",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-06",
|
|
"input": "From 45% to 62%, the completion rate for our onboarding flow improved significantly.",
|
|
"responseDurationMs": 1,
|
|
"actualPrimaryType": "observed_problem",
|
|
"actualReasoningModes": [
|
|
"identify_difference"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": false,
|
|
"reasoningModeMatch": false,
|
|
"nextQuestionPresent": true,
|
|
"pass": false,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "completion rate",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "45%",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "62%",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "onboarding",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "all improvements are due to the redesign",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "the old flow was bad",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "users prefer the new design",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-07",
|
|
"input": "A user claimed that our pricing model is too complex for small businesses.",
|
|
"responseDurationMs": 1,
|
|
"actualPrimaryType": "reported_claim",
|
|
"actualReasoningModes": [
|
|
"identify_difference",
|
|
"validate_claim"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": true,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": true,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "pricing complexity",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "small business",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "user claim",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "the pricing is actually complex",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "other small businesses agree",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "we should simplify pricing",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-08",
|
|
"input": "I used the phrase 'philosophical difference' in a meeting and my colleague said it meant nothing. Is that fair?",
|
|
"responseDurationMs": 1,
|
|
"actualPrimaryType": "ambiguous_statement",
|
|
"actualReasoningModes": [
|
|
"identify_difference",
|
|
"clarify_meaning"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": true,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": true,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "philosophical",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "ambiguous",
|
|
"found": true
|
|
},
|
|
{
|
|
"concept": "meaning clarification",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "the phrase was wrong",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "the colleague is hostile",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "we should avoid philosophical language",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-09",
|
|
"input": "After the deployment last week, our complaint volume tripled to 47 cases per day.",
|
|
"responseDurationMs": 1,
|
|
"actualPrimaryType": "causal_claim",
|
|
"actualReasoningModes": [
|
|
"establish_baseline",
|
|
"identify_difference"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": true,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": true,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "deployment",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "complaint volume increase",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "tripled",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "47 cases",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "the deployment caused the complaints",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "the bug report was insufficient",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "rollback is needed",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
},
|
|
{
|
|
"id": "diag-10",
|
|
"input": "Some complaints involve production issues, but others say the delivery team is slow.",
|
|
"responseDurationMs": 1,
|
|
"actualPrimaryType": "observed_problem",
|
|
"actualReasoningModes": [
|
|
"establish_baseline",
|
|
"identify_difference"
|
|
],
|
|
"technical": {
|
|
"schemaValid": true,
|
|
"classificationMatch": true,
|
|
"reasoningModeMatch": true,
|
|
"nextQuestionPresent": true,
|
|
"pass": true,
|
|
"errors": []
|
|
},
|
|
"reasoningQuality": {
|
|
"requiredConcepts": {
|
|
"pass": false,
|
|
"details": [
|
|
{
|
|
"concept": "production issues",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "delivery speed",
|
|
"found": false
|
|
},
|
|
{
|
|
"concept": "complaint types",
|
|
"found": false
|
|
}
|
|
]
|
|
},
|
|
"unsupportedInferencesAbsent": {
|
|
"pass": true,
|
|
"details": [
|
|
{
|
|
"concept": "production is worse than delivery",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "the delivery team needs training",
|
|
"absent": true
|
|
},
|
|
{
|
|
"concept": "both teams are underperforming equally",
|
|
"absent": true
|
|
}
|
|
]
|
|
},
|
|
"pass": false
|
|
}
|
|
}
|
|
]
|
|
} |