forge-workflow 0.0.3 → 0.0.5

package/lib/dep-guard/path-utils.js

@@ -1,13 +1,13 @@
-const path = require('node:path');
-
-function toPosixPath(filePath) {
-	return filePath.replaceAll('\\', '/').split(path.sep).join('/');
-}
-
-function normalizeRepoPath(filePath) {
-	return toPosixPath(path.normalize(filePath));
-}
-
-module.exports = {
-	normalizeRepoPath,
-};
+const path = require('node:path');
+
+function toPosixPath(filePath) {
+	return filePath.replaceAll('\\', '/').split(path.sep).join('/');
+}
+
+function normalizeRepoPath(filePath) {
+	return toPosixPath(path.normalize(filePath));
+}
+
+module.exports = {
+	normalizeRepoPath,
+};

package/lib/dep-guard/rubric.js

@@ -1,120 +1,120 @@
-const WEIGHTS = {
-	importCallChain: 3,
-	contractDependencies: 3,
-	behavioralDependencies: 2,
-};
-
-const CONFLICT_MESSAGES = {
-	materialDisagreement: 'Detector scores disagree materially across dependency categories.',
-	behavioralUncertainty: 'Behavioral signals are heuristic and need manual review.',
-};
-const MAX_DETECTOR_STRENGTH = 3;
-const MAX_TOTAL = Object.values(WEIGHTS)
-	.reduce((sum, weight) => sum + (weight * MAX_DETECTOR_STRENGTH), 0);
-
-function aggregateRubricScores(detectors, options = {}) {
-	const confidenceThreshold = options.confidenceThreshold ?? 0.7;
-	const dimensions = {};
-	let total = 0;
-	let hasMaterialDisagreement = false;
-	let hasUncertainty = false;
-
-	for (const [name, weight] of Object.entries(WEIGHTS)) {
-		const raw = detectors[name]?.score ?? 0;
-		const normalized = Math.min(raw, MAX_DETECTOR_STRENGTH);
-		const weighted = normalized * weight;
-		dimensions[name] = { raw, normalized, weighted };
-		total += weighted;
-		if (detectors[name]?.uncertain) {
-			hasUncertainty = true;
-		}
-	}
-
-	if (total === 0) {
-		return {
-			result: 'PASS',
-			score: 0,
-			total: 0,
-			dimensions,
-			weights: { ...WEIGHTS },
-			confidence: {
-				score: 1,
-				threshold: confidenceThreshold,
-				belowThreshold: false,
-				reasons: [],
-			},
-			detectorConflicts: [],
-			summary: 'No detector findings yet.',
-			reasons: [],
-		};
-	}
-
-	const rawScores = Object.values(dimensions).map((dimension) => dimension.raw);
-	const nonZeroScores = rawScores.filter((score) => score > 0);
-	const detectorConflicts = [];
-
-	if (nonZeroScores.length >= 2 && (Math.max(...nonZeroScores) - Math.min(...nonZeroScores) >= 2)) {
-		hasMaterialDisagreement = true;
-		detectorConflicts.push(CONFLICT_MESSAGES.materialDisagreement);
-	}
-
-	if (hasUncertainty) {
-		detectorConflicts.push(CONFLICT_MESSAGES.behavioralUncertainty);
-	}
-
-	let confidenceScore = total / MAX_TOTAL;
-	if (hasMaterialDisagreement) {
-		confidenceScore -= 0.2;
-	}
-	if (hasUncertainty) {
-		confidenceScore -= 0.15;
-	}
-	confidenceScore = Math.max(0, Math.min(1, confidenceScore));
-	const belowThreshold = confidenceScore < confidenceThreshold;
-
-	const reasons = [];
-	if (detectorConflicts.length > 0) {
-		reasons.push(...detectorConflicts);
-	}
-	if (belowThreshold) {
-		reasons.push(`Confidence ${confidenceScore.toFixed(2)} is below the ${(confidenceThreshold * 100).toFixed(0)}% threshold.`);
-	}
-
-	let result = 'PASS';
-	if (detectorConflicts.length > 0) {
-		result = 'INCONCLUSIVE';
-	} else if (belowThreshold) {
-		result = 'WEAK';
-	}
-
-	return {
-		result,
-		score: total,
-		total,
-		dimensions,
-		weights: { ...WEIGHTS },
-		confidence: {
-			score: confidenceScore,
-			threshold: confidenceThreshold,
-			belowThreshold,
-			reasons,
-		},
-		detectorConflicts,
-		summary: reasons.join(' ') || 'Detector findings are consistent.',
-		reasons,
-	};
-}
-
-function needsUserEscalation(rubricResult) {
-	return (
-		rubricResult.result === 'INCONCLUSIVE'
-		|| rubricResult.confidence.belowThreshold
-		|| rubricResult.detectorConflicts.length > 0
-	);
-}
-
-module.exports = {
-	WEIGHTS,
-	aggregateRubricScores,
-	needsUserEscalation,
-};
+const WEIGHTS = {
+	importCallChain: 3,
+	contractDependencies: 3,
+	behavioralDependencies: 2,
+};
+
+const CONFLICT_MESSAGES = {
+	materialDisagreement: 'Detector scores disagree materially across dependency categories.',
+	behavioralUncertainty: 'Behavioral signals are heuristic and need manual review.',
+};
+const MAX_DETECTOR_STRENGTH = 3;
+const MAX_TOTAL = Object.values(WEIGHTS)
+	.reduce((sum, weight) => sum + (weight * MAX_DETECTOR_STRENGTH), 0);
+
+function aggregateRubricScores(detectors, options = {}) {
+	const confidenceThreshold = options.confidenceThreshold ?? 0.7;
+	const dimensions = {};
+	let total = 0;
+	let hasMaterialDisagreement = false;
+	let hasUncertainty = false;
+
+	for (const [name, weight] of Object.entries(WEIGHTS)) {
+		const raw = detectors[name]?.score ?? 0;
+		const normalized = Math.min(raw, MAX_DETECTOR_STRENGTH);
+		const weighted = normalized * weight;
+		dimensions[name] = { raw, normalized, weighted };
+		total += weighted;
+		if (detectors[name]?.uncertain) {
+			hasUncertainty = true;
+		}
+	}
+
+	if (total === 0) {
+		return {
+			result: 'PASS',
+			score: 0,
+			total: 0,
+			dimensions,
+			weights: { ...WEIGHTS },
+			confidence: {
+				score: 1,
+				threshold: confidenceThreshold,
+				belowThreshold: false,
+				reasons: [],
+			},
+			detectorConflicts: [],
+			summary: 'No detector findings yet.',
+			reasons: [],
+		};
+	}
+
+	const rawScores = Object.values(dimensions).map((dimension) => dimension.raw);
+	const nonZeroScores = rawScores.filter((score) => score > 0);
+	const detectorConflicts = [];
+
+	if (nonZeroScores.length >= 2 && (Math.max(...nonZeroScores) - Math.min(...nonZeroScores) >= 2)) {
+		hasMaterialDisagreement = true;
+		detectorConflicts.push(CONFLICT_MESSAGES.materialDisagreement);
+	}
+
+	if (hasUncertainty) {
+		detectorConflicts.push(CONFLICT_MESSAGES.behavioralUncertainty);
+	}
+
+	let confidenceScore = total / MAX_TOTAL;
+	if (hasMaterialDisagreement) {
+		confidenceScore -= 0.2;
+	}
+	if (hasUncertainty) {
+		confidenceScore -= 0.15;
+	}
+	confidenceScore = Math.max(0, Math.min(1, confidenceScore));
+	const belowThreshold = confidenceScore < confidenceThreshold;
+
+	const reasons = [];
+	if (detectorConflicts.length > 0) {
+		reasons.push(...detectorConflicts);
+	}
+	if (belowThreshold) {
+		reasons.push(`Confidence ${confidenceScore.toFixed(2)} is below the ${(confidenceThreshold * 100).toFixed(0)}% threshold.`);
+	}
+
+	let result = 'PASS';
+	if (detectorConflicts.length > 0) {
+		result = 'INCONCLUSIVE';
+	} else if (belowThreshold) {
+		result = 'WEAK';
+	}
+
+	return {
+		result,
+		score: total,
+		total,
+		dimensions,
+		weights: { ...WEIGHTS },
+		confidence: {
+			score: confidenceScore,
+			threshold: confidenceThreshold,
+			belowThreshold,
+			reasons,
+		},
+		detectorConflicts,
+		summary: reasons.join(' ') || 'Detector findings are consistent.',
+		reasons,
+	};
+}
+
+function needsUserEscalation(rubricResult) {
+	return (
+		rubricResult.result === 'INCONCLUSIVE'
+		|| rubricResult.confidence.belowThreshold
+		|| rubricResult.detectorConflicts.length > 0
+	);
+}
+
+module.exports = {
+	WEIGHTS,
+	aggregateRubricScores,
+	needsUserEscalation,
+};