@dotsetlabs/bellwether 1.0.3 → 2.0.0

package/dist/baseline/diff.js

@@ -73,16 +73,15 @@ export function formatDiffText(diff, useColors = true) {
         lines.push(red('─── Performance Regressions ───'));
         for (const regression of diff.performanceReport.regressions) {
             const percentStr = (regression.regressionPercent * 100).toFixed(1);
-            const confidenceNote = regression.isReliable
-                ? ''
-                : ` ${yellow('(low confidence)')}`;
+            const confidenceNote = regression.isReliable ? '' : ` ${yellow('(low confidence)')}`;
             lines.push(`  ${red('!')} ${regression.toolName}: ` +
                 `${regression.previousP50Ms.toFixed(0)}ms → ` +
                 `${regression.currentP50Ms.toFixed(0)}ms (+${percentStr}%)${confidenceNote}`);
         }
         lines.push('');
         // Show low confidence tools warning
-        if (diff.performanceReport.lowConfidenceTools && diff.performanceReport.lowConfidenceTools.length > 0) {
+        if (diff.performanceReport.lowConfidenceTools &&
+            diff.performanceReport.lowConfidenceTools.length > 0) {
             lines.push(yellow('  Note: Some tools have low confidence metrics.'));
             lines.push(yellow(`  Run with more samples for reliable baselines: ${diff.performanceReport.lowConfidenceTools.join(', ')}`));
             lines.push('');
@@ -94,7 +93,8 @@ export function formatDiffText(diff, useColors = true) {
         lines.push('');
     }
     // Performance confidence changes
-    if (diff.performanceReport?.confidenceChanges && diff.performanceReport.confidenceChanges.length > 0) {
+    if (diff.performanceReport?.confidenceChanges &&
+        diff.performanceReport.confidenceChanges.length > 0) {
         lines.push(cyan('─── Confidence Changes ───'));
         for (const change of diff.performanceReport.confidenceChanges) {
             const icon = change.improved ? green('↑') : change.degraded ? yellow('↓') : '→';
@@ -212,7 +212,8 @@ export function formatDiffText(diff, useColors = true) {
     lines.push(`  Info: ${diff.infoCount}`);
     if (diff.performanceReport) {
         lines.push(`  Performance regressions: ${diff.performanceReport.regressionCount}`);
-        if (diff.performanceReport.lowConfidenceTools && diff.performanceReport.lowConfidenceTools.length > 0) {
+        if (diff.performanceReport.lowConfidenceTools &&
+            diff.performanceReport.lowConfidenceTools.length > 0) {
             lines.push(`  Low confidence tools: ${diff.performanceReport.lowConfidenceTools.length}`);
         }
     }
@@ -276,7 +277,8 @@ export function formatDiffCompact(diff) {
     if (diff.performanceReport?.regressionCount ?? 0 > 0) {
         parts.push(`perf_regressions=${diff.performanceReport?.regressionCount}`);
     }
-    if (diff.performanceReport?.lowConfidenceTools && diff.performanceReport.lowConfidenceTools.length > 0) {
+    if (diff.performanceReport?.lowConfidenceTools &&
+        diff.performanceReport.lowConfidenceTools.length > 0) {
         parts.push(`low_confidence_tools=${diff.performanceReport.lowConfidenceTools.length}`);
     }
     if (diff.securityReport) {
@@ -336,8 +338,11 @@ export function formatDiffGitHubActions(diff) {
         lines.push(`::notice::Minor changes: ${diff.summary}`);
     }
     for (const change of diff.behaviorChanges) {
-        const level = change.severity === 'breaking' ? 'error' :
-            change.severity === 'warning' ? 'warning' : 'notice';
+        const level = change.severity === 'breaking'
+            ? 'error'
+            : change.severity === 'warning'
+                ? 'warning'
+                : 'notice';
         lines.push(`::${level}::${change.tool} - ${change.description}`);
     }
     for (const tool of diff.toolsRemoved) {
@@ -355,7 +360,8 @@ export function formatDiffGitHubActions(diff) {
         }
     }
     // Low confidence warning
-    if (diff.performanceReport?.lowConfidenceTools && diff.performanceReport.lowConfidenceTools.length > 0) {
+    if (diff.performanceReport?.lowConfidenceTools &&
+        diff.performanceReport.lowConfidenceTools.length > 0) {
         lines.push(`::notice::Low confidence metrics for ${diff.performanceReport.lowConfidenceTools.length} tool(s): ${diff.performanceReport.lowConfidenceTools.join(', ')}`);
     }
     // Security findings
@@ -440,7 +446,9 @@ export function formatDiffMarkdown(diff) {
                 toolDiff.schemaChanged ? 'Schema changed' : '',
                 toolDiff.descriptionChanged ? 'Description changed' : '',
                 `${toolDiff.changes.length} change(s)`,
-            ].filter(Boolean).join(', ');
+            ]
+                .filter(Boolean)
+                .join(', ');
             lines.push(`| ${toolDiff.tool} | ⚠️ Modified | ${details} |`);
         }
         lines.push('');
@@ -451,8 +459,7 @@ export function formatDiffMarkdown(diff) {
         lines.push('| Tool | Aspect | Severity | Description |');
         lines.push('|------|--------|----------|-------------|');
         for (const change of diff.behaviorChanges) {
-            const sevEmoji = change.severity === 'breaking' ? '🔴' :
-                change.severity === 'warning' ? '🟡' : '🟢';
+            const sevEmoji = change.severity === 'breaking' ? '🔴' : change.severity === 'warning' ? '🟡' : '🟢';
             lines.push(`| ${change.tool} | ${change.aspect} | ${sevEmoji} ${change.severity} | ${change.description} |`);
         }
         lines.push('');
@@ -503,7 +510,11 @@ export function formatDiffMarkdown(diff) {
                 lines.push('|------|--------|---------|');
                 for (const issue of schemaReport.toolsWithIssues) {
                     const statusIcon = issue.isBreaking ? '🔴' : issue.becameUnstable ? '🟡' : '🔵';
-                    const status = issue.isBreaking ? 'Breaking' : issue.becameUnstable ? 'Unstable' : 'Changed';
+                    const status = issue.isBreaking
+                        ? 'Breaking'
+                        : issue.becameUnstable
+                            ? 'Unstable'
+                            : 'Changed';
                     lines.push(`| ${issue.toolName} | ${statusIcon} ${status} | ${issue.summary} |`);
                 }
                 lines.push('');
@@ -525,11 +536,13 @@ export function formatDiffMarkdown(diff) {
                 lines.push(`⚠️ **Error behavior changed**: ${et.summary}`);
                 lines.push('');
             }
-            if (et.newCategories.length > 0 || et.resolvedCategories.length > 0 ||
-                et.increasingCategories.length > 0 || et.decreasingCategories.length > 0) {
+            if (et.newCategories.length > 0 ||
+                et.resolvedCategories.length > 0 ||
+                et.increasingCategories.length > 0 ||
+                et.decreasingCategories.length > 0) {
                 lines.push('| Category | Trend | Previous | Current | Change |');
                 lines.push('|----------|-------|----------|---------|--------|');
-                for (const trend of et.trends.filter(t => t.trend !== 'stable')) {
+                for (const trend of et.trends.filter((t) => t.trend !== 'stable')) {
                     const trendEmoji = getTrendEmoji(trend.trend);
                     const changeStr = trend.changePercent !== 0
                         ? `${trend.changePercent > 0 ? '+' : ''}${trend.changePercent}%`
@@ -543,7 +556,8 @@ export function formatDiffMarkdown(diff) {
     // Performance section
     if (diff.performanceReport) {
         const perfReport = diff.performanceReport;
-        if (perfReport.hasRegressions || perfReport.improvementCount > 0 ||
+        if (perfReport.hasRegressions ||
+            perfReport.improvementCount > 0 ||
             (perfReport.lowConfidenceTools && perfReport.lowConfidenceTools.length > 0)) {
             lines.push('### Performance');
             lines.push('');
@@ -605,7 +619,8 @@ export function formatDiffMarkdown(diff) {
     lines.push(`- Info: **${diff.infoCount}**`);
     if (diff.performanceReport) {
         lines.push(`- Performance regressions: **${diff.performanceReport.regressionCount}**`);
-        if (diff.performanceReport.lowConfidenceTools && diff.performanceReport.lowConfidenceTools.length > 0) {
+        if (diff.performanceReport.lowConfidenceTools &&
+            diff.performanceReport.lowConfidenceTools.length > 0) {
             lines.push(`- Low confidence tools: **${diff.performanceReport.lowConfidenceTools.length}**`);
         }
     }
@@ -674,9 +689,7 @@ function getTrendEmoji(trend) {
  */
 export function formatDiffJUnit(diff, suiteName = 'bellwether') {
     const timestamp = new Date().toISOString();
-    const totalTests = diff.toolsAdded.length +
-        diff.toolsRemoved.length +
-        diff.behaviorChanges.length;
+    const totalTests = diff.toolsAdded.length + diff.toolsRemoved.length + diff.behaviorChanges.length;
     const failures = diff.breakingCount;
     const errors = 0;
     const skipped = 0;
@@ -743,7 +756,8 @@ export function formatDiffJUnit(diff, suiteName = 'bellwether') {
             lines.push('    </testcase>');
         }
         // Low confidence tools
-        if (diff.performanceReport.lowConfidenceTools && diff.performanceReport.lowConfidenceTools.length > 0) {
+        if (diff.performanceReport.lowConfidenceTools &&
+            diff.performanceReport.lowConfidenceTools.length > 0) {
             for (const tool of diff.performanceReport.lowConfidenceTools) {
                 const name = escapeXml(`confidence-${tool}`);
                 lines.push(`    <testcase name="${name}" classname="drift.confidence">`);
@@ -805,7 +819,8 @@ export function formatDiffJUnit(diff, suiteName = 'bellwether') {
             lines.push('    </testcase>');
         }
         // Show stable schemas as passing tests
-        if (diff.schemaEvolutionReport.stableCount > 0 && diff.schemaEvolutionReport.toolsWithIssues.length === 0) {
+        if (diff.schemaEvolutionReport.stableCount > 0 &&
+            diff.schemaEvolutionReport.toolsWithIssues.length === 0) {
             lines.push(`    <testcase name="schema-stability-check" classname="drift.schema">`);
             lines.push(`      <system-out>${diff.schemaEvolutionReport.stableCount} tool(s) have stable response schemas</system-out>`);
             lines.push('    </testcase>');
@@ -830,7 +845,7 @@ export function formatDiffJUnit(diff, suiteName = 'bellwether') {
         }
         // Increasing error types (warnings)
         for (const category of et.increasingCategories) {
-            const trend = et.trends.find(t => t.category === category);
+            const trend = et.trends.find((t) => t.category === category);
             const name = escapeXml(`error-trend-increasing-${category}`);
             lines.push(`    <testcase name="${name}" classname="drift.errors">`);
             lines.push(`      <system-err>[WARNING] Error frequency increasing: ${escapeXml(category)}${trend ? ` (+${trend.changePercent}%)` : ''}</system-err>`);
@@ -878,7 +893,7 @@ export function formatDiffJUnit(diff, suiteName = 'bellwether') {
  * Format diff as SARIF (Static Analysis Results Interchange Format) for GitHub Code Scanning.
  *
  * SARIF is the standard format for GitHub's code scanning feature and can be
- * uploaded to show drift detection results in pull request reviews.
+ * used to show drift detection results in pull request reviews.
  *
  * @see https://docs.oasis-open.org/sarif/sarif/v2.1.0/sarif-v2.1.0.html
  *
@@ -1313,11 +1328,13 @@ export function formatDiffSarif(diff, baselinePath = 'bellwether-baseline.json')
         }
         // Increasing error types
         for (const category of et.increasingCategories) {
-            const trend = et.trends.find(t => t.category === category);
+            const trend = et.trends.find((t) => t.category === category);
             results.push({
                 ruleId: 'BWH011',
                 level: 'warning',
-                message: { text: `Error frequency increasing: ${category}${trend ? ` (+${trend.changePercent}%)` : ''}` },
+                message: {
+                    text: `Error frequency increasing: ${category}${trend ? ` (+${trend.changePercent}%)` : ''}`,
+                },
                 locations: [
                     {
                         physicalLocation: {

package/dist/baseline/error-analyzer.d.ts

@@ -69,7 +69,7 @@ export interface ErrorAnalysisSummary {
     /** Unique remediations suggested */
     remediations: string[];
     /** Counts by error category */
-    categoryCounts: Map<string, number>;
+    categoryCounts: Map<string, number> | Record<string, number>;
     /** Top root causes (most common) */
     topRootCauses: string[];
     /** Top remediations (most actionable) */

package/dist/baseline/error-analyzer.js

@@ -22,7 +22,8 @@ export function analyzeError(errorMessage, context) {
     let statusCategory = categorizeHttpStatus(httpStatus);
     const wasExpected = context?.wasExpected ?? context?.expectedOutcome === 'error';
     // If the error was expected (validation test), recategorize it
-    if (wasExpected && (statusCategory === 'client_error_validation' || statusCategory === 'unknown')) {
+    if (wasExpected &&
+        (statusCategory === 'client_error_validation' || statusCategory === 'unknown')) {
         statusCategory = 'validation_expected';
     }
     const rootCause = wasExpected
@@ -96,7 +97,9 @@ export function generateErrorSummary(toolName, patterns) {
         }
     }
     // Count transient and actionable errors
-    const transientErrors = analyses.filter((a) => a.transient).reduce((sum, a) => sum + a.pattern.count, 0);
+    const transientErrors = analyses
+        .filter((a) => a.transient)
+        .reduce((sum, a) => sum + a.pattern.count, 0);
     const actionableCount = analyses.filter((a) => a.remediation && !a.remediation.includes('Review')).length;
     // Collect unique remediations with frequency
     const remediationCounts = new Map();
@@ -324,7 +327,9 @@ export function inferRootCause(message, category) {
     if (lower.includes('invalid') || lower.includes('malformed')) {
         return 'Invalid input format or value';
     }
-    if (lower.includes('not found') || lower.includes('does not exist') || lower.includes("doesn't exist")) {
+    if (lower.includes('not found') ||
+        lower.includes('does not exist') ||
+        lower.includes("doesn't exist")) {
         return 'Referenced resource does not exist';
     }
     if (lower.includes('already exists') || lower.includes('duplicate')) {
@@ -333,7 +338,9 @@ export function inferRootCause(message, category) {
     if (lower.includes('unauthorized') || lower.includes('authentication')) {
         return 'Authentication credentials missing or invalid';
     }
-    if (lower.includes('forbidden') || lower.includes('permission') || lower.includes('access denied')) {
+    if (lower.includes('forbidden') ||
+        lower.includes('permission') ||
+        lower.includes('access denied')) {
         return 'Insufficient permissions for this operation';
     }
     if (lower.includes('rate') || lower.includes('throttl') || lower.includes('too many')) {
@@ -480,15 +487,69 @@ export function extractRelatedParameters(message) {
  */
 function isCommonWord(word) {
     const commonWords = new Set([
-        'the', 'is', 'are', 'was', 'were', 'be', 'been', 'being',
-        'have', 'has', 'had', 'do', 'does', 'did', 'will', 'would',
-        'could', 'should', 'may', 'might', 'must', 'shall', 'can',
-        'need', 'not', 'and', 'but', 'or', 'if', 'then', 'else',
-        'for', 'with', 'from', 'this', 'that', 'these', 'those',
-        'error', 'message', 'failed', 'invalid', 'missing', 'required',
-        'found', 'exist', 'exists', 'value', 'input', 'output', 'type',
-        'string', 'number', 'boolean', 'object', 'array', 'null', 'undefined',
-        'field', 'parameter', 'property', 'argument', 'key',
+        'the',
+        'is',
+        'are',
+        'was',
+        'were',
+        'be',
+        'been',
+        'being',
+        'have',
+        'has',
+        'had',
+        'do',
+        'does',
+        'did',
+        'will',
+        'would',
+        'could',
+        'should',
+        'may',
+        'might',
+        'must',
+        'shall',
+        'can',
+        'need',
+        'not',
+        'and',
+        'but',
+        'or',
+        'if',
+        'then',
+        'else',
+        'for',
+        'with',
+        'from',
+        'this',
+        'that',
+        'these',
+        'those',
+        'error',
+        'message',
+        'failed',
+        'invalid',
+        'missing',
+        'required',
+        'found',
+        'exist',
+        'exists',
+        'value',
+        'input',
+        'output',
+        'type',
+        'string',
+        'number',
+        'boolean',
+        'object',
+        'array',
+        'null',
+        'undefined',
+        'field',
+        'parameter',
+        'property',
+        'argument',
+        'key',
     ]);
     return commonWords.has(word.toLowerCase());
 }
@@ -510,9 +571,19 @@ export function isTransientError(category, message) {
         return true;
     // Check for transient keywords
     const transientKeywords = [
-        'timeout', 'timed out', 'temporarily', 'retry', 'unavailable',
-        'connection', 'network', 'service unavailable', 'too many requests',
-        'try again', 'overloaded', 'busy', 'maintenance',
+        'timeout',
+        'timed out',
+        'temporarily',
+        'retry',
+        'unavailable',
+        'connection',
+        'network',
+        'service unavailable',
+        'too many requests',
+        'try again',
+        'overloaded',
+        'busy',
+        'maintenance',
     ];
     return transientKeywords.some((keyword) => lower.includes(keyword));
 }
@@ -661,7 +732,9 @@ export function formatErrorTrendReport(report, useColors = false) {
     lines.push('  Trend details:');
     for (const trend of report.trends.filter((t) => t.trend !== 'stable')) {
         const arrow = getTrendArrow(trend.trend);
-        const changeText = trend.changePercent !== 0 ? ` (${trend.changePercent > 0 ? '+' : ''}${trend.changePercent}%)` : '';
+        const changeText = trend.changePercent !== 0
+            ? ` (${trend.changePercent > 0 ? '+' : ''}${trend.changePercent}%)`
+            : '';
         lines.push(`    ${arrow} ${trend.category}: ${trend.previousCount} → ${trend.currentCount}${changeText}`);
     }
     return lines.join('\n');

package/dist/baseline/incremental-checker.js

@@ -26,7 +26,7 @@ export function analyzeForIncremental(currentTools, baseline, options = {}) {
     // If no baseline or force retest, test everything
     if (!baseline || forceRetest) {
         return {
-            toolsToTest: currentTools.map(t => t.name),
+            toolsToTest: currentTools.map((t) => t.name),
             toolsToSkip: [],
             cachedFingerprints: [],
             changeSummary: {
@@ -34,7 +34,7 @@ export function analyzeForIncremental(currentTools, baseline, options = {}) {
                 changedTools: 0,
                 unchangedTools: 0,
                 removedTools: 0,
-                newToolNames: baseline ? [] : currentTools.map(t => t.name),
+                newToolNames: baseline ? [] : currentTools.map((t) => t.name),
                 changedToolNames: [],
                 removedToolNames: [],
             },
@@ -47,8 +47,8 @@ export function analyzeForIncremental(currentTools, baseline, options = {}) {
     const changedToolNames = [];
     const removedToolNames = [];
     // Build maps for comparison
-    const baselineToolMap = new Map(getToolFingerprints(baseline).map(t => [t.name, t]));
-    const currentToolSet = new Set(currentTools.map(t => t.name));
+    const baselineToolMap = new Map(getToolFingerprints(baseline).map((t) => [t.name, t]));
+    const currentToolSet = new Set(currentTools.map((t) => t.name));
     // Check current tools against baseline
     for (const tool of currentTools) {
         const baselineTool = baselineToolMap.get(tool.name);
@@ -65,7 +65,10 @@ export function analyzeForIncremental(currentTools, baseline, options = {}) {
         }
         // Check if schema changed
         const currentSchemaHash = computeSchemaHash(tool.inputSchema);
-        const baselineSchemaHash = baselineTool.schemaHash;
+        const baselineSchemaHash = baselineTool.inputSchemaHashAtTest ??
+            (baselineTool.inputSchema
+                ? computeSchemaHash(baselineTool.inputSchema)
+                : baselineTool.schemaHash);
         if (currentSchemaHash !== baselineSchemaHash) {
             // Schema changed - needs retesting
             toolsToTest.push(tool.name);

package/dist/baseline/index.d.ts

@@ -6,26 +6,16 @@ export { getBaselineGeneratedAt, getBaselineHash, getBaselineServerCommand, getB
 export { createBaseline, saveBaseline, loadBaseline, verifyBaselineHash, baselineExists, recalculateBaselineHash, acceptDrift, hasAcceptance, clearAcceptance, type LoadBaselineOptions, type AcceptDriftOptions, } from './saver.js';
 export { compareWithBaseline, compareBaselines, hasBreakingChanges, hasSecurityChanges, filterByMinimumSeverity, checkBaselineVersionCompatibility, compareSeverity, severityMeetsThreshold, applyAspectOverride, applySeverityConfig, shouldFailOnDiff, } from './comparator.js';
 export { formatDiffText, formatDiffJson, formatDiffCompact, formatDiffGitHubActions, formatDiffMarkdown, formatDiffJUnit, formatDiffSarif, formatSecurityReport, } from './diff.js';
-export { createCloudBaseline, } from './converter.js';
+export { createBaselineFromInterview } from './converter.js';
 export { computeSchemaHash, compareSchemas, computeConsensusSchemaHash, type SchemaChangeType, type SchemaChange, type SchemaComparisonResult, } from './schema-compare.js';
 export { getBaselineVersion, parseVersion, areVersionsCompatible, compareVersions, getCompatibilityWarning, checkVersionCompatibility, assertVersionCompatibility, formatVersion, isCurrentVersion, isOlderVersion, isNewerVersion, requiresMigration, BaselineVersionError, type FormatVersion, type VersionCompatibility, } from './version.js';
 export { analyzeForIncremental, mergeFingerprints, formatIncrementalSummary, isIncrementalWorthwhile, addIncrementalMetadata, type IncrementalCheckResult, type IncrementalChangeSummary, type IncrementalCheckOptions, } from './incremental-checker.js';
 export { analyzeResponses, inferSchemaFromValue, compareFingerprints, compareErrorPatterns, computeInferredSchemaHash, type ResponseFingerprint, type ResponseContentType, type ResponseSize, type InferredSchema, type ErrorPattern, type ResponseAnalysis, type FingerprintDiff, type FingerprintChange, type ErrorPatternDiff, } from './response-fingerprint.js';
 export { compareInferredSchemas, buildSchemaEvolution, compareSchemaEvolution, formatSchemaEvolution, formatSchemaEvolutionDiff, hasSchemaEvolutionIssues, getSchemaStabilityGrade, type ResponseSchemaEvolution, type SchemaVersion as SchemaEvolutionVersion, type SchemaEvolutionDiff, type SchemaTypeChange, } from './response-schema-tracker.js';
-export { analyzeToolChangeImpact, analyzeDiffImpact, analyzeSchemaChanges, isBreakingChange, getBreakingChangeSummary, CHANGE_IMPACT, type SchemaChangeType as ImpactSchemaChangeType, type SchemaChangeDetail, type MigrationComplexity, type ChangeImpact, type DiffImpactAnalysis, type ActionItem, } from './change-impact-analyzer.js';
-export { calculateMetrics, createPerformanceBaseline, extractPerformanceBaselines, comparePerformance, generatePerformanceReport, formatMetrics, formatComparison, isPerformanceAcceptable, aggregateSamplesByTool, calculatePerformanceConfidence, calculateConfidenceFromMetrics, formatConfidenceLevel, hasReliableConfidence, PERFORMANCE, type LatencyTrend, type ToolPerformanceMetrics, type PerformanceBaseline, type PerformanceComparison, type PerformanceReport, type LatencySample, } from './performance-tracker.js';
-export { checkDeprecations, checkToolDeprecation, markAsDeprecated, clearDeprecation, getDeprecatedTools, getExpiredTools, getUpcomingRemovals, formatDeprecationWarning, formatDeprecationReport, shouldFailOnDeprecation, DEPRECATION, DEPRECATION_DEFAULTS, DEPRECATION_THRESHOLDS, type DeprecationStatus, type DeprecationWarning, type DeprecationReport, type DeprecationConfig, } from './deprecation-tracker.js';
-export { calculateHealthScore, formatHealthScore, meetsHealthThreshold, getHealthBadgeColor, createHealthHistoryEntry, HEALTH_SCORING, HEALTH_WEIGHTS, GRADE_THRESHOLDS, SEVERITY_THRESHOLDS, HEALTH_PENALTIES, type HealthTrend, type ActionPriority, type HealthActionItem, type HealthComponents, type HealthScore, type HealthHistory, type HealthInput, } from './health-scorer.js';
-export { buildServerTimeline, buildToolTimeline, formatTimeline, formatServerTimelineSummary, generateVisualTimeline, serializeTimeline, deserializeTimeline, serializeServerTimeline, deserializeServerTimeline, getMostActiveTools, getMostBreakingTools, getBreakingChanges, getVersionAtTime, getChangesBetween, hadBreakingChanges, type SchemaEventType, type SchemaVersion, type SchemaTimeline, type ServerTimeline, type DeprecationEvent, type TimelineStats, type TimelineBuildOptions, } from './schema-evolution.js';
-export { generateMigrationGuide, formatMigrationGuideMarkdown, formatMigrationGuideText, hasBreakingMigrationChanges, getBreakingTools, type MigrationEffort, type MigrationStepType, type CodeExample, type BreakingChange, type MigrationStep, type MigrationGuide, type MigrationStats, } from './migration-generator.js';
-export { generateToolScenarios, generateBaselineScenarios, formatScenariosAsYaml, formatScenariosReport, getScenariosByPriority, getScenariosByCategory, getCriticalScenarios, getSecurityScenarios, type ScenarioCategory, type ScenarioPriority, type TestScenario, type AutoGeneratedScenarios, type ScenarioGenerationSummary, type ScenarioGenerationResult, type ScenarioGenerationConfig, } from './scenario-generator.js';
-export { generatePRComment, generateCompactPRComment, generateCIStatusSummary, generateDiffTable, generateBadgeUrl, generateBadgeMarkdown, getBadgeColor, shouldBlockMerge, getSeverityEmoji, type BadgeColor, type CommentSection, type AffectedWorkflow, type PRComment, type PRCommentConfig, } from './pr-comment-generator.js';
+export { calculateMetrics, createPerformanceBaseline, extractPerformanceBaselines, comparePerformance, generatePerformanceReport, formatMetrics, formatComparison, isPerformanceAcceptable, aggregateSamplesByTool, calculatePerformanceConfidence, calculateConfidenceFromMetrics, formatConfidenceLevel, hasReliableConfidence, type LatencyTrend, type ToolPerformanceMetrics, type PerformanceBaseline, type PerformanceComparison, type PerformanceReport, type LatencySample, } from './performance-tracker.js';
 export type { SecurityCategory, RiskLevel, SecurityPayload, SecurityTestResult, SecurityFinding, SecurityFingerprint, SecurityDiff, SecurityTestOptions, SecurityTestContext, SecurityToolCallResult, SecurityReport, } from '../security/types.js';
 export { runSecurityTests, compareSecurityFingerprints, getRiskLevelFromScore, parseSecurityCategories, getPayloadsForCategory, getAllSecurityPayloads, getAllSecurityCategories, } from '../security/index.js';
 export type { HttpStatusCategory, ErrorSeverity, EnhancedErrorAnalysis, ErrorAnalysisSummary, ErrorTrend, ErrorTrendReport, } from './error-analyzer.js';
 export { analyzeError, analyzeErrorPatterns, generateErrorSummary, analyzeErrorTrends, extractHttpStatus, categorizeHttpStatus, inferRootCause, generateRemediation, extractRelatedParameters, isTransientError, assessErrorSeverity, mapStatusToErrorCategory, formatEnhancedError, formatErrorTrendReport, formatCategoryName, } from './error-analyzer.js';
 export { scoreDocumentation, scoreToolDocumentation, calculateDescriptionCoverage, calculateDescriptionQuality, calculateParameterDocumentation, calculateExampleCoverage, hasExamples, scoreToGrade, generateSuggestions, compareDocumentationScores, formatDocumentationScore, formatDocumentationScoreCompact, formatDocumentationScoreChange, toDocumentationScoreSummary, getGradeIndicator, getGradeBadgeColor, meetsDocumentationThreshold, meetsDocumentationGrade, type DocumentationIssueSeverity, type DocumentationIssueType, } from './documentation-scorer.js';
-export { calculateAICompatibilityScore, generateAICompatibilityMarkdown, type AICompatibilityScore, type ScoreComponent, type AICompatibilityRecommendation, type ToolAIScore, type AICompatibilityInput, } from './ai-compatibility-scorer.js';
-export { calculateRiskScore, generateRiskScoreMarkdown, type RegressionRiskScore, type RiskFactor, } from './risk-scorer.js';
-export { calculatePruningDecisions, calculateToolPruning, prioritizeTools, generatePruningSummary, generatePruningMarkdown, type TestCategory, type TestCategoryDecision, type ToolPruningDecision, type ToolCharacteristics, type PruningInput, type PruningSummary, } from './test-pruner.js';
 //# sourceMappingURL=index.d.ts.map

package/dist/baseline/index.js

@@ -5,39 +5,19 @@ export { getBaselineGeneratedAt, getBaselineHash, getBaselineServerCommand, getB
 export { createBaseline, saveBaseline, loadBaseline, verifyBaselineHash, baselineExists, recalculateBaselineHash, acceptDrift, hasAcceptance, clearAcceptance, } from './saver.js';
 export { compareWithBaseline, compareBaselines, hasBreakingChanges, hasSecurityChanges, filterByMinimumSeverity, checkBaselineVersionCompatibility, compareSeverity, severityMeetsThreshold, applyAspectOverride, applySeverityConfig, shouldFailOnDiff, } from './comparator.js';
 export { formatDiffText, formatDiffJson, formatDiffCompact, formatDiffGitHubActions, formatDiffMarkdown, formatDiffJUnit, formatDiffSarif, formatSecurityReport, } from './diff.js';
-export { createCloudBaseline, } from './converter.js';
+export { createBaselineFromInterview } from './converter.js';
 export { computeSchemaHash, compareSchemas, computeConsensusSchemaHash, } from './schema-compare.js';
 export { getBaselineVersion, parseVersion, areVersionsCompatible, compareVersions, getCompatibilityWarning, checkVersionCompatibility, assertVersionCompatibility, formatVersion, isCurrentVersion, isOlderVersion, isNewerVersion, requiresMigration, BaselineVersionError, } from './version.js';
-// Legacy baseline migrations removed; cloud baseline is canonical.
+// Legacy baseline migrations removed; current baseline format is canonical.
 // Incremental checking
 export { analyzeForIncremental, mergeFingerprints, formatIncrementalSummary, isIncrementalWorthwhile, addIncrementalMetadata, } from './incremental-checker.js';
 export { analyzeResponses, inferSchemaFromValue, compareFingerprints, compareErrorPatterns, computeInferredSchemaHash, } from './response-fingerprint.js';
 // Response schema evolution tracking
 export { compareInferredSchemas, buildSchemaEvolution, compareSchemaEvolution, formatSchemaEvolution, formatSchemaEvolutionDiff, hasSchemaEvolutionIssues, getSchemaStabilityGrade, } from './response-schema-tracker.js';
-// Change impact analysis
-export { analyzeToolChangeImpact, analyzeDiffImpact, analyzeSchemaChanges, isBreakingChange, getBreakingChangeSummary, CHANGE_IMPACT, } from './change-impact-analyzer.js';
 // Performance tracking
-export { calculateMetrics, createPerformanceBaseline, extractPerformanceBaselines, comparePerformance, generatePerformanceReport, formatMetrics, formatComparison, isPerformanceAcceptable, aggregateSamplesByTool, calculatePerformanceConfidence, calculateConfidenceFromMetrics, formatConfidenceLevel, hasReliableConfidence, PERFORMANCE, } from './performance-tracker.js';
-// Deprecation tracking
-export { checkDeprecations, checkToolDeprecation, markAsDeprecated, clearDeprecation, getDeprecatedTools, getExpiredTools, getUpcomingRemovals, formatDeprecationWarning, formatDeprecationReport, shouldFailOnDeprecation, DEPRECATION, DEPRECATION_DEFAULTS, DEPRECATION_THRESHOLDS, } from './deprecation-tracker.js';
-// Health scoring
-export { calculateHealthScore, formatHealthScore, meetsHealthThreshold, getHealthBadgeColor, createHealthHistoryEntry, HEALTH_SCORING, HEALTH_WEIGHTS, GRADE_THRESHOLDS, SEVERITY_THRESHOLDS, HEALTH_PENALTIES, } from './health-scorer.js';
-// Schema evolution timeline
-export { buildServerTimeline, buildToolTimeline, formatTimeline, formatServerTimelineSummary, generateVisualTimeline, serializeTimeline, deserializeTimeline, serializeServerTimeline, deserializeServerTimeline, getMostActiveTools, getMostBreakingTools, getBreakingChanges, getVersionAtTime, getChangesBetween, hadBreakingChanges, } from './schema-evolution.js';
-// Migration guide generation
-export { generateMigrationGuide, formatMigrationGuideMarkdown, formatMigrationGuideText, hasBreakingMigrationChanges, getBreakingTools, } from './migration-generator.js';
-// Auto-generated test scenarios
-export { generateToolScenarios, generateBaselineScenarios, formatScenariosAsYaml, formatScenariosReport, getScenariosByPriority, getScenariosByCategory, getCriticalScenarios, getSecurityScenarios, } from './scenario-generator.js';
-// Enhanced PR comments
-export { generatePRComment, generateCompactPRComment, generateCIStatusSummary, generateDiffTable, generateBadgeUrl, generateBadgeMarkdown, getBadgeColor, shouldBlockMerge, getSeverityEmoji, } from './pr-comment-generator.js';
+export { calculateMetrics, createPerformanceBaseline, extractPerformanceBaselines, comparePerformance, generatePerformanceReport, formatMetrics, formatComparison, isPerformanceAcceptable, aggregateSamplesByTool, calculatePerformanceConfidence, calculateConfidenceFromMetrics, formatConfidenceLevel, hasReliableConfidence, } from './performance-tracker.js';
 export { runSecurityTests, compareSecurityFingerprints, getRiskLevelFromScore, parseSecurityCategories, getPayloadsForCategory, getAllSecurityPayloads, getAllSecurityCategories, } from '../security/index.js';
 export { analyzeError, analyzeErrorPatterns, generateErrorSummary, analyzeErrorTrends, extractHttpStatus, categorizeHttpStatus, inferRootCause, generateRemediation, extractRelatedParameters, isTransientError, assessErrorSeverity, mapStatusToErrorCategory, formatEnhancedError, formatErrorTrendReport, formatCategoryName, } from './error-analyzer.js';
 // Documentation quality scoring
 export { scoreDocumentation, scoreToolDocumentation, calculateDescriptionCoverage, calculateDescriptionQuality, calculateParameterDocumentation, calculateExampleCoverage, hasExamples, scoreToGrade, generateSuggestions, compareDocumentationScores, formatDocumentationScore, formatDocumentationScoreCompact, formatDocumentationScoreChange, toDocumentationScoreSummary, getGradeIndicator, getGradeBadgeColor, meetsDocumentationThreshold, meetsDocumentationGrade, } from './documentation-scorer.js';
-// AI Agent Compatibility Scoring
-export { calculateAICompatibilityScore, generateAICompatibilityMarkdown, } from './ai-compatibility-scorer.js';
-// Regression Risk Scoring
-export { calculateRiskScore, generateRiskScoreMarkdown, } from './risk-scorer.js';
-// Intelligent Test Pruning
-export { calculatePruningDecisions, calculateToolPruning, prioritizeTools, generatePruningSummary, generatePruningMarkdown, } from './test-pruner.js';
 //# sourceMappingURL=index.js.map

package/dist/baseline/performance-tracker.d.ts

@@ -132,7 +132,6 @@ export interface LatencySample {
      */
     outcomeCorrect?: boolean;
 }
-export { PERFORMANCE_TRACKING as PERFORMANCE } from '../constants.js';
 /**
  * Calculate statistical confidence for performance metrics.
  *

package/dist/baseline/performance-tracker.js

@@ -6,8 +6,6 @@
  */
 import { getBaselineGeneratedAt, getToolFingerprints } from './accessors.js';
 import { PERFORMANCE_TRACKING, PERFORMANCE_CONFIDENCE } from '../constants.js';
-// Re-export centralized constant for backwards compatibility
-export { PERFORMANCE_TRACKING as PERFORMANCE } from '../constants.js';
 /**
  * Calculate statistical confidence for performance metrics.
  *
@@ -45,14 +43,14 @@ export function calculatePerformanceConfidence(samples, options = {}) {
     }
     // Categorize samples by expected outcome
     // Happy path tests: expectedOutcome === 'success' or undefined (backward compat)
-    const happyPathSamples = samples.filter(s => s.expectedOutcome === 'success' || s.expectedOutcome === undefined);
+    const happyPathSamples = samples.filter((s) => s.expectedOutcome === 'success' || s.expectedOutcome === undefined);
     // Validation tests: expectedOutcome === 'error'
-    const validationTestSamples = samples.filter(s => s.expectedOutcome === 'error');
+    const validationTestSamples = samples.filter((s) => s.expectedOutcome === 'error');
     // Count validation samples that correctly rejected (error as expected = success)
-    const validationSuccesses = validationTestSamples.filter(s => !s.success && (s.outcomeCorrect === undefined || s.outcomeCorrect === true)).length;
+    const validationSuccesses = validationTestSamples.filter((s) => !s.success && (s.outcomeCorrect === undefined || s.outcomeCorrect === true)).length;
     // For confidence, only use happy path samples that succeeded
-    const successfulHappyPath = happyPathSamples.filter(s => s.success);
-    const allDurations = successfulHappyPath.map(s => s.durationMs);
+    const successfulHappyPath = happyPathSamples.filter((s) => s.success);
+    const allDurations = successfulHappyPath.map((s) => s.durationMs);
     // Handle all failures case
     if (allDurations.length === 0) {
         return {
@@ -69,12 +67,10 @@ export function calculatePerformanceConfidence(samples, options = {}) {
     // For variance calculation, exclude the first sample (cold start warmup)
     // This prevents JIT compilation, connection setup, and cache warming from
     // inflating the coefficient of variation and lowering confidence scores.
-    const durationsForVariance = excludeWarmup && allDurations.length > 1
-        ? allDurations.slice(1)
-        : allDurations;
+    const durationsForVariance = excludeWarmup && allDurations.length > 1 ? allDurations.slice(1) : allDurations;
     // Calculate variance using post-warmup samples only
     const meanForVariance = durationsForVariance.reduce((sum, d) => sum + d, 0) / durationsForVariance.length;
-    const squaredDiffs = durationsForVariance.map(d => Math.pow(d - meanForVariance, 2));
+    const squaredDiffs = durationsForVariance.map((d) => Math.pow(d - meanForVariance, 2));
     const variance = squaredDiffs.reduce((sum, d) => sum + d, 0) / durationsForVariance.length;
     const standardDeviation = Math.sqrt(variance);
     // Calculate coefficient of variation (CV = stdDev / mean)
@@ -186,8 +182,8 @@ export function calculateMetrics(samples) {
         return null;
     }
     const toolName = samples[0].toolName;
-    const successfulSamples = samples.filter(s => s.success);
-    const durations = successfulSamples.map(s => s.durationMs).sort((a, b) => a - b);
+    const successfulSamples = samples.filter((s) => s.success);
+    const durations = successfulSamples.map((s) => s.durationMs).sort((a, b) => a - b);
     if (durations.length === 0) {
         // All calls failed
         const confidence = calculatePerformanceConfidence(samples);
@@ -213,7 +209,7 @@ export function calculateMetrics(samples) {
     const minMs = durations[0];
     const maxMs = durations[durations.length - 1];
     // Calculate standard deviation
-    const squaredDiffs = durations.map(d => Math.pow(d - avgMs, 2));
+    const squaredDiffs = durations.map((d) => Math.pow(d - avgMs, 2));
     const avgSquaredDiff = squaredDiffs.reduce((sum, d) => sum + d, 0) / squaredDiffs.length;
     const stdDevMs = Math.sqrt(avgSquaredDiff);
     // Calculate confidence from samples
@@ -323,8 +319,8 @@ export function comparePerformance(current, baseline, regressionThreshold = PERF
     const trend = determineTrend(p50Regression);
     // Check for regression
     const maxRegression = baseline.maxAllowedRegression ?? regressionThreshold;
-    const hasRegression = p50Regression !== null && p50Regression > maxRegression ||
-        p95Regression !== null && p95Regression > maxRegression;
+    const hasRegression = (p50Regression !== null && p50Regression > maxRegression) ||
+        (p95Regression !== null && p95Regression > maxRegression);
     // Determine severity
     const severity = determinePerformanceSeverity(p50Regression, p95Regression, maxRegression);
     // Generate summary (include confidence note if low)
@@ -520,10 +516,7 @@ export function formatMetrics(metrics) {
  * Format performance comparison for display.
  */
 export function formatComparison(comparison) {
-    const lines = [
-        `Tool: ${comparison.toolName}`,
-        `  Trend: ${comparison.trend.toUpperCase()}`,
-    ];
+    const lines = [`Tool: ${comparison.toolName}`, `  Trend: ${comparison.trend.toUpperCase()}`];
     if (comparison.p50RegressionPercent !== null) {
         const sign = comparison.p50RegressionPercent >= 0 ? '+' : '';
         lines.push(`  p50 change: ${sign}${(comparison.p50RegressionPercent * 100).toFixed(1)}%`);