@llm-dev-ops/agentics-cli 1.4.27 → 1.4.29

package/dist/bundled-agents/copilot-agents/lib/services/agents/meta-reasoner/meta-reasoner-agent.js

@@ -1,686 +0,0 @@
-"use strict";
-/**
- * Meta-Reasoner Agent
- *
- * Purpose: Evaluate reasoning quality and consistency across agents
- * Classification: META_ANALYSIS, REASONING_QUALITY_ASSESSMENT
- * decision_type: meta_reasoning_analysis
- *
- * Scope:
- * - Detect contradictions
- * - Assess confidence calibration
- * - Identify systemic reasoning issues
- *
- * CONSTITUTION COMPLIANCE:
- * ✓ Stateless at runtime
- * ✓ Emits exactly ONE DecisionEvent per invocation
- * ✓ Persists ONLY via ruvector-service
- * ✓ NEVER connects directly to databases
- * ✓ NEVER executes SQL
- * ✓ NEVER modifies runtime behavior
- * ✓ NEVER orchestrates other agents
- * ✓ NEVER enforces policy
- * ✓ NEVER intercepts execution paths
- *
- * Must Never:
- * - Override outputs
- * - Enforce corrections
- * - Execute logic
- */
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.MetaReasonerAgent = void 0;
-const uuid_1 = require("uuid");
-const contracts_1 = require("../contracts");
-const AGENT_ID = 'meta-reasoner-agent';
-const AGENT_VERSION = '1.0.0';
-const DECISION_TYPE = 'meta_reasoning_analysis';
-/**
- * Meta-Reasoner Agent Implementation
- *
- * This agent analyzes reasoning traces from other agents to detect
- * contradictions, assess confidence calibration, and identify systemic issues.
- * It is purely analytical - it NEVER overrides outputs or enforces corrections.
- */
-class MetaReasonerAgent {
-    metadata = {
-        id: AGENT_ID,
-        name: 'Meta-Reasoner Agent',
-        version: AGENT_VERSION,
-        classifications: [
-            contracts_1.AgentClassification.META_ANALYSIS,
-            contracts_1.AgentClassification.REASONING_QUALITY_ASSESSMENT,
-        ],
-        decision_type: DECISION_TYPE,
-        description: 'Evaluates reasoning quality and consistency across agents by detecting contradictions, assessing confidence calibration, and identifying systemic reasoning issues.',
-    };
-    persistence;
-    telemetry;
-    constructor(persistence, telemetry) {
-        this.persistence = persistence;
-        this.telemetry = telemetry;
-    }
-    /**
-     * Validate input against MetaReasonerInputSchema
-     */
-    validateInput(input) {
-        return contracts_1.MetaReasonerInputSchema.parse(input);
-    }
-    /**
-     * Invoke the meta-reasoner agent
-     *
-     * DETERMINISTIC: Same input always produces same output structure
-     * STATELESS: No internal state modified
-     * NON-BLOCKING: Fully async
-     */
-    async invoke(input, executionRef) {
-        const startTime = Date.now();
-        try {
-            // Emit telemetry start
-            this.telemetry.recordStart(AGENT_ID, executionRef, input);
-            // Perform meta-reasoning analysis (pure analysis, no side effects)
-            const output = this.analyzeReasoning(input);
-            // Validate output
-            const validatedOutput = contracts_1.MetaReasonerOutputSchema.parse(output);
-            // Calculate confidence based on analysis quality
-            const confidence = this.calculateConfidence(validatedOutput, input);
-            // Constraints applied during analysis
-            const constraintsApplied = this.getAppliedConstraints(input);
-            // Create the DecisionEvent
-            const event = (0, contracts_1.createDecisionEvent)(AGENT_ID, AGENT_VERSION, DECISION_TYPE, input, validatedOutput, confidence, constraintsApplied, executionRef);
-            // Persist via ruvector-service (best-effort, non-blocking)
-            let persistence_status;
-            try {
-                await this.persistence.store(event);
-                persistence_status = { status: 'persisted' };
-            }
-            catch (persistError) {
-                const persistMessage = persistError instanceof Error ? persistError.message : 'Unknown persistence error';
-                console.error(`[${AGENT_ID}] RuVector persistence failed (non-blocking): ${persistMessage}`);
-                persistence_status = { status: 'skipped', error: persistMessage };
-            }
-            // Emit telemetry success
-            this.telemetry.recordSuccess(AGENT_ID, executionRef, Date.now() - startTime);
-            return {
-                status: 'success',
-                event,
-                persistence_status,
-            };
-        }
-        catch (error) {
-            const errorMessage = error instanceof Error ? error.message : 'Unknown error';
-            const errorCode = this.classifyError(error);
-            // Emit telemetry failure
-            this.telemetry.recordFailure(AGENT_ID, executionRef, errorCode, errorMessage);
-            return (0, contracts_1.createErrorResult)(errorCode, errorMessage, executionRef);
-        }
-    }
-    /**
-     * Perform meta-reasoning analysis on the provided traces
-     *
-     * This is the core analysis logic - purely analytical.
-     * NEVER overrides outputs or enforces corrections.
-     */
-    analyzeReasoning(input) {
-        const analysisId = (0, uuid_1.v4)();
-        const traces = input.traces;
-        const scope = input.scope;
-        // Collect unique agents and decision types
-        const uniqueAgents = new Set(traces.map(t => t.agent_id));
-        const uniqueDecisionTypes = new Set(traces.map(t => t.decision_type));
-        // Calculate time span
-        const timestamps = traces.map(t => new Date(t.timestamp).getTime()).sort((a, b) => a - b);
-        const timeSpan = timestamps.length > 0 ? {
-            earliest: new Date(timestamps[0]).toISOString(),
-            latest: new Date(timestamps[timestamps.length - 1]).toISOString(),
-        } : undefined;
-        // Perform analyses based on scope
-        const contradictions = scope.detect_contradictions
-            ? this.detectContradictions(traces)
-            : [];
-        const confidenceCalibrations = scope.assess_confidence_calibration
-            ? this.assessConfidenceCalibration(traces, input.context?.historical_accuracy)
-            : [];
-        const systemicIssues = scope.identify_systemic_issues
-            ? this.identifySystemicIssues(traces)
-            : [];
-        // Calculate quality metrics
-        const qualityMetrics = this.calculateQualityMetrics(traces, contradictions, confidenceCalibrations, systemicIssues);
-        // Generate key findings
-        const keyFindings = this.generateKeyFindings(contradictions, confidenceCalibrations, systemicIssues, qualityMetrics);
-        // Generate summary
-        const summary = this.generateSummary(traces.length, uniqueAgents.size, contradictions.length, systemicIssues.length, qualityMetrics.overall_score);
-        // Extract assumptions
-        const assumptions = this.extractAssumptions(input);
-        return {
-            analysis_id: analysisId,
-            summary,
-            quality_metrics: qualityMetrics,
-            contradictions,
-            confidence_calibrations: confidenceCalibrations,
-            systemic_issues: systemicIssues,
-            analysis_metadata: {
-                total_traces: traces.length,
-                traces_analyzed: traces.length,
-                unique_agents: uniqueAgents.size,
-                unique_decision_types: uniqueDecisionTypes.size,
-                time_span: timeSpan,
-                scope_executed: {
-                    contradictions_checked: scope.detect_contradictions,
-                    calibration_assessed: scope.assess_confidence_calibration,
-                    systemic_issues_checked: scope.identify_systemic_issues,
-                    fallacies_checked: scope.detect_fallacies,
-                    completeness_checked: scope.check_completeness,
-                },
-            },
-            key_findings: keyFindings,
-            assumptions,
-            version: '1.0.0',
-        };
-    }
-    /**
-     * Detect contradictions between reasoning traces
-     */
-    detectContradictions(traces) {
-        const contradictions = [];
-        let contradictionCount = 0;
-        // Group traces by decision type for comparison
-        const tracesByType = new Map();
-        for (const trace of traces) {
-            const existing = tracesByType.get(trace.decision_type) || [];
-            existing.push(trace);
-            tracesByType.set(trace.decision_type, existing);
-        }
-        // Check for contradictions within same decision type
-        for (const [decisionType, typeTraces] of tracesByType) {
-            if (typeTraces.length < 2)
-                continue;
-            // Compare pairs of traces
-            for (let i = 0; i < typeTraces.length; i++) {
-                for (let j = i + 1; j < typeTraces.length; j++) {
-                    const trace1 = typeTraces[i];
-                    const trace2 = typeTraces[j];
-                    // Check for confidence contradictions
-                    const confidenceGap = Math.abs(trace1.reported_confidence - trace2.reported_confidence);
-                    if (confidenceGap > 0.5) {
-                        contradictionCount++;
-                        contradictions.push({
-                            contradiction_id: `contradiction-${contradictionCount}`,
-                            type: 'statistical',
-                            severity: confidenceGap > 0.7 ? 'high' : 'medium',
-                            involved_traces: [trace1.execution_ref, trace2.execution_ref],
-                            involved_agents: [trace1.agent_id, trace2.agent_id],
-                            description: `Large confidence gap (${(confidenceGap * 100).toFixed(1)}%) between agents for same decision type "${decisionType}"`,
-                            evidence: [
-                                {
-                                    trace_ref: trace1.execution_ref,
-                                    excerpt: `Confidence: ${trace1.reported_confidence}`,
-                                    relevance: 'First trace confidence value',
-                                },
-                                {
-                                    trace_ref: trace2.execution_ref,
-                                    excerpt: `Confidence: ${trace2.reported_confidence}`,
-                                    relevance: 'Second trace confidence value',
-                                },
-                            ],
-                            finding_confidence: 0.8,
-                        });
-                    }
-                    // Check for constraint contradictions
-                    const constraints1 = new Set(trace1.constraints_applied);
-                    const constraints2 = new Set(trace2.constraints_applied);
-                    const conflictingConstraints = this.findConflictingConstraints(constraints1, constraints2);
-                    if (conflictingConstraints.length > 0) {
-                        contradictionCount++;
-                        contradictions.push({
-                            contradiction_id: `contradiction-${contradictionCount}`,
-                            type: 'contextual',
-                            severity: 'medium',
-                            involved_traces: [trace1.execution_ref, trace2.execution_ref],
-                            involved_agents: [trace1.agent_id, trace2.agent_id],
-                            description: `Conflicting constraints applied for same decision type "${decisionType}"`,
-                            evidence: [
-                                {
-                                    trace_ref: trace1.execution_ref,
-                                    excerpt: `Constraints: ${Array.from(constraints1).join(', ')}`,
-                                    relevance: 'First trace constraints',
-                                },
-                                {
-                                    trace_ref: trace2.execution_ref,
-                                    excerpt: `Constraints: ${Array.from(constraints2).join(', ')}`,
-                                    relevance: 'Second trace constraints',
-                                },
-                            ],
-                            finding_confidence: 0.7,
-                        });
-                    }
-                }
-            }
-        }
-        // Check for temporal contradictions (same agent, changing conclusions)
-        const tracesByAgent = new Map();
-        for (const trace of traces) {
-            const existing = tracesByAgent.get(trace.agent_id) || [];
-            existing.push(trace);
-            tracesByAgent.set(trace.agent_id, existing);
-        }
-        for (const [agentId, agentTraces] of tracesByAgent) {
-            if (agentTraces.length < 2)
-                continue;
-            // Sort by timestamp
-            const sorted = [...agentTraces].sort((a, b) => new Date(a.timestamp).getTime() - new Date(b.timestamp).getTime());
-            // Check for rapid confidence fluctuation
-            for (let i = 1; i < sorted.length; i++) {
-                const prev = sorted[i - 1];
-                const curr = sorted[i];
-                if (prev.decision_type === curr.decision_type) {
-                    const confidenceChange = Math.abs(curr.reported_confidence - prev.reported_confidence);
-                    if (confidenceChange > 0.3) {
-                        contradictionCount++;
-                        contradictions.push({
-                            contradiction_id: `contradiction-${contradictionCount}`,
-                            type: 'temporal',
-                            severity: confidenceChange > 0.5 ? 'high' : 'low',
-                            involved_traces: [prev.execution_ref, curr.execution_ref],
-                            involved_agents: [agentId],
-                            description: `Rapid confidence change (${(confidenceChange * 100).toFixed(1)}%) for agent "${agentId}" on same decision type`,
-                            evidence: [
-                                {
-                                    trace_ref: prev.execution_ref,
-                                    excerpt: `Earlier confidence: ${prev.reported_confidence} at ${prev.timestamp}`,
-                                    relevance: 'Previous confidence value',
-                                },
-                                {
-                                    trace_ref: curr.execution_ref,
-                                    excerpt: `Later confidence: ${curr.reported_confidence} at ${curr.timestamp}`,
-                                    relevance: 'Current confidence value',
-                                },
-                            ],
-                            finding_confidence: 0.75,
-                        });
-                    }
-                }
-            }
-        }
-        return contradictions;
-    }
-    /**
-     * Find conflicting constraints between two sets
-     */
-    findConflictingConstraints(set1, set2) {
-        const conflicts = [];
-        // Define known conflicting constraint pairs
-        const conflictPairs = [
-            ['strict_validation', 'relaxed_validation'],
-            ['high_precision', 'high_recall'],
-            ['fast_execution', 'thorough_analysis'],
-        ];
-        for (const [a, b] of conflictPairs) {
-            if ((set1.has(a) && set2.has(b)) || (set1.has(b) && set2.has(a))) {
-                conflicts.push(`${a} vs ${b}`);
-            }
-        }
-        return conflicts;
-    }
-    /**
-     * Assess confidence calibration for each agent
-     */
-    assessConfidenceCalibration(traces, historicalAccuracy) {
-        const calibrations = [];
-        // Group traces by agent
-        const tracesByAgent = new Map();
-        for (const trace of traces) {
-            const existing = tracesByAgent.get(trace.agent_id) || [];
-            existing.push(trace);
-            tracesByAgent.set(trace.agent_id, existing);
-        }
-        for (const [agentId, agentTraces] of tracesByAgent) {
-            const confidences = agentTraces.map(t => t.reported_confidence);
-            const meanConfidence = confidences.reduce((a, b) => a + b, 0) / confidences.length;
-            const expectedAccuracy = historicalAccuracy?.[agentId];
-            let calibrationScore;
-            let assessment;
-            let calibrationGap;
-            const recommendations = [];
-            if (expectedAccuracy !== undefined) {
-                calibrationGap = meanConfidence - expectedAccuracy;
-                if (Math.abs(calibrationGap) <= 0.1) {
-                    calibrationScore = 0.9;
-                    assessment = 'well_calibrated';
-                }
-                else if (calibrationGap > 0.1) {
-                    calibrationScore = Math.max(0, 1 - calibrationGap);
-                    assessment = 'overconfident';
-                    recommendations.push(`Consider reducing confidence by ~${(calibrationGap * 100).toFixed(0)}%`, 'Review historical accuracy data for calibration');
-                }
-                else {
-                    calibrationScore = Math.max(0, 1 + calibrationGap);
-                    assessment = 'underconfident';
-                    recommendations.push(`Consider increasing confidence by ~${(Math.abs(calibrationGap) * 100).toFixed(0)}%`, 'Agent may be underestimating its capabilities');
-                }
-            }
-            else {
-                // Without historical data, assess based on consistency
-                const confidenceVariance = this.calculateVariance(confidences);
-                if (confidenceVariance > 0.1) {
-                    calibrationScore = Math.max(0, 1 - confidenceVariance);
-                    assessment = 'inconsistent';
-                    recommendations.push('Confidence varies significantly across invocations', 'Consider implementing more stable confidence estimation');
-                }
-                else {
-                    calibrationScore = 0.7; // Default moderate score without historical data
-                    assessment = 'insufficient_data';
-                    recommendations.push('Historical accuracy data needed for proper calibration assessment', 'Track prediction outcomes to build accuracy baseline');
-                }
-            }
-            calibrations.push({
-                agent_id: agentId,
-                calibration_score: calibrationScore,
-                assessment,
-                mean_reported_confidence: meanConfidence,
-                expected_accuracy: expectedAccuracy,
-                calibration_gap: calibrationGap,
-                traces_analyzed: agentTraces.length,
-                recommendations,
-            });
-        }
-        return calibrations;
-    }
-    /**
-     * Calculate variance of an array of numbers
-     */
-    calculateVariance(values) {
-        if (values.length < 2)
-            return 0;
-        const mean = values.reduce((a, b) => a + b, 0) / values.length;
-        const squaredDiffs = values.map(v => Math.pow(v - mean, 2));
-        return squaredDiffs.reduce((a, b) => a + b, 0) / values.length;
-    }
-    /**
-     * Identify systemic reasoning issues across traces
-     */
-    identifySystemicIssues(traces) {
-        const issues = [];
-        let issueCount = 0;
-        // Check for anchoring bias: early traces heavily influencing later ones
-        const sortedTraces = [...traces].sort((a, b) => new Date(a.timestamp).getTime() - new Date(b.timestamp).getTime());
-        if (sortedTraces.length >= 3) {
-            const earlyConfidence = sortedTraces.slice(0, Math.floor(sortedTraces.length / 3))
-                .map(t => t.reported_confidence);
-            const lateConfidence = sortedTraces.slice(-Math.floor(sortedTraces.length / 3))
-                .map(t => t.reported_confidence);
-            const earlyMean = earlyConfidence.reduce((a, b) => a + b, 0) / earlyConfidence.length;
-            const lateMean = lateConfidence.reduce((a, b) => a + b, 0) / lateConfidence.length;
-            if (Math.abs(earlyMean - lateMean) < 0.05 && this.calculateVariance(earlyConfidence) < 0.01) {
-                issueCount++;
-                issues.push({
-                    issue_id: `issue-${issueCount}`,
-                    type: 'anchoring_bias',
-                    severity: 'medium',
-                    affected_agents: [...new Set(traces.map(t => t.agent_id))],
-                    occurrences: sortedTraces.slice(0, 3).map(t => t.execution_ref),
-                    frequency: 'occasional',
-                    description: 'Early confidence values may be anchoring subsequent analyses',
-                    evidence: [
-                        {
-                            trace_ref: sortedTraces[0].execution_ref,
-                            observation: `Early mean confidence: ${earlyMean.toFixed(3)}`,
-                        },
-                        {
-                            trace_ref: sortedTraces[sortedTraces.length - 1].execution_ref,
-                            observation: `Late mean confidence: ${lateMean.toFixed(3)} (minimal drift suggests anchoring)`,
-                        },
-                    ],
-                    impact: 'May reduce adaptability to new information',
-                    finding_confidence: 0.6,
-                });
-            }
-        }
-        // Check for missing uncertainty acknowledgment
-        const highConfidenceTraces = traces.filter(t => t.reported_confidence > 0.9);
-        if (highConfidenceTraces.length > traces.length * 0.5) {
-            issueCount++;
-            issues.push({
-                issue_id: `issue-${issueCount}`,
-                type: 'missing_uncertainty',
-                severity: 'medium',
-                affected_agents: [...new Set(highConfidenceTraces.map(t => t.agent_id))],
-                occurrences: highConfidenceTraces.slice(0, 5).map(t => t.execution_ref),
-                frequency: highConfidenceTraces.length > traces.length * 0.7 ? 'pervasive' : 'frequent',
-                description: 'Majority of traces report very high confidence (>90%), potentially underestimating uncertainty',
-                evidence: highConfidenceTraces.slice(0, 3).map(t => ({
-                    trace_ref: t.execution_ref,
-                    observation: `Confidence: ${(t.reported_confidence * 100).toFixed(1)}%`,
-                })),
-                impact: 'High confidence without uncertainty bounds may lead to overreliance on outputs',
-                finding_confidence: 0.75,
-            });
-        }
-        // Check for inconsistent criteria across agents
-        const constraintSets = new Map();
-        for (const trace of traces) {
-            const existing = constraintSets.get(trace.decision_type) || new Set();
-            trace.constraints_applied.forEach(c => existing.add(c));
-            constraintSets.set(trace.decision_type, existing);
-        }
-        for (const [decisionType, constraints] of constraintSets) {
-            const tracesOfType = traces.filter(t => t.decision_type === decisionType);
-            const uniqueConstraintCombinations = new Set(tracesOfType.map(t => JSON.stringify([...t.constraints_applied].sort())));
-            if (uniqueConstraintCombinations.size > 3 && tracesOfType.length > 5) {
-                issueCount++;
-                issues.push({
-                    issue_id: `issue-${issueCount}`,
-                    type: 'inconsistent_criteria',
-                    severity: 'low',
-                    affected_agents: [...new Set(tracesOfType.map(t => t.agent_id))],
-                    occurrences: tracesOfType.slice(0, 3).map(t => t.execution_ref),
-                    frequency: 'occasional',
-                    description: `Multiple different constraint sets (${uniqueConstraintCombinations.size}) used for decision type "${decisionType}"`,
-                    evidence: tracesOfType.slice(0, 2).map(t => ({
-                        trace_ref: t.execution_ref,
-                        observation: `Constraints: ${t.constraints_applied.join(', ')}`,
-                    })),
-                    impact: 'May indicate inconsistent evaluation standards',
-                    finding_confidence: 0.65,
-                });
-            }
-        }
-        return issues;
-    }
-    /**
-     * Calculate overall reasoning quality metrics
-     */
-    calculateQualityMetrics(traces, contradictions, calibrations, systemicIssues) {
-        // Consistency score: penalize for contradictions
-        const contradictionPenalty = Math.min(contradictions.length * 0.1, 0.5);
-        const consistencyScore = Math.max(0, 1 - contradictionPenalty);
-        // Completeness score: based on traces having required fields
-        const tracesWithConstraints = traces.filter(t => t.constraints_applied.length > 0).length;
-        const completenessScore = traces.length > 0 ? tracesWithConstraints / traces.length : 0;
-        // Clarity score: based on calibration quality
-        const avgCalibrationScore = calibrations.length > 0
-            ? calibrations.reduce((a, b) => a + b.calibration_score, 0) / calibrations.length
-            : 0.7;
-        const clarityScore = avgCalibrationScore;
-        // Constraint adherence score
-        const constraintAdherenceScore = traces.length > 0 ? completenessScore : 0;
-        // Calculate overall score
-        const overallScore = (consistencyScore * 0.35 +
-            completenessScore * 0.2 +
-            clarityScore * 0.25 +
-            constraintAdherenceScore * 0.2) * (1 - Math.min(systemicIssues.length * 0.05, 0.3));
-        const uniqueAgents = new Set(traces.map(t => t.agent_id));
-        return {
-            overall_score: Math.max(0, Math.min(1, overallScore)),
-            consistency_score: consistencyScore,
-            completeness_score: completenessScore,
-            clarity_score: clarityScore,
-            constraint_adherence_score: constraintAdherenceScore,
-            traces_analyzed: traces.length,
-            agents_analyzed: uniqueAgents.size,
-            coverage_percentage: 100, // All provided traces are analyzed
-        };
-    }
-    /**
-     * Generate prioritized key findings
-     */
-    generateKeyFindings(contradictions, calibrations, systemicIssues, qualityMetrics) {
-        const findings = [];
-        // Critical/high severity contradictions
-        const criticalContradictions = contradictions.filter(c => c.severity === 'critical' || c.severity === 'high');
-        if (criticalContradictions.length > 0) {
-            findings.push({
-                priority: 1,
-                category: 'contradiction',
-                finding: `Found ${criticalContradictions.length} high-severity contradiction(s) requiring attention`,
-                affected_entities: criticalContradictions.flatMap(c => c.involved_agents),
-            });
-        }
-        // Calibration issues
-        const poorlyCalibrated = calibrations.filter(c => c.assessment === 'overconfident' || c.assessment === 'underconfident');
-        if (poorlyCalibrated.length > 0) {
-            findings.push({
-                priority: 2,
-                category: 'calibration',
-                finding: `${poorlyCalibrated.length} agent(s) show calibration issues (${poorlyCalibrated.map(c => c.assessment).join(', ')})`,
-                affected_entities: poorlyCalibrated.map(c => c.agent_id),
-            });
-        }
-        // Systemic issues
-        const severeSystems = systemicIssues.filter(i => i.severity === 'critical' || i.severity === 'high');
-        if (severeSystems.length > 0) {
-            findings.push({
-                priority: 3,
-                category: 'systemic',
-                finding: `Identified ${severeSystems.length} systemic issue(s): ${severeSystems.map(i => i.type).join(', ')}`,
-                affected_entities: severeSystems.flatMap(i => i.affected_agents),
-            });
-        }
-        // Overall quality assessment
-        if (qualityMetrics.overall_score < 0.5) {
-            findings.push({
-                priority: 4,
-                category: 'quality',
-                finding: `Overall reasoning quality score is low (${(qualityMetrics.overall_score * 100).toFixed(1)}%)`,
-                affected_entities: [],
-            });
-        }
-        else if (qualityMetrics.overall_score >= 0.8) {
-            findings.push({
-                priority: 10,
-                category: 'quality',
-                finding: `Overall reasoning quality is good (${(qualityMetrics.overall_score * 100).toFixed(1)}%)`,
-                affected_entities: [],
-            });
-        }
-        // Sort by priority
-        return findings.sort((a, b) => a.priority - b.priority);
-    }
-    /**
-     * Generate summary of the analysis
-     */
-    generateSummary(traceCount, agentCount, contradictionCount, issueCount, overallScore) {
-        const qualityDescriptor = overallScore >= 0.8 ? 'good' :
-            overallScore >= 0.6 ? 'moderate' :
-                overallScore >= 0.4 ? 'fair' : 'poor';
-        let summary = `Meta-reasoning analysis of ${traceCount} trace(s) from ${agentCount} agent(s). `;
-        summary += `Overall reasoning quality is ${qualityDescriptor} (${(overallScore * 100).toFixed(1)}%). `;
-        if (contradictionCount > 0) {
-            summary += `Detected ${contradictionCount} contradiction(s). `;
-        }
-        if (issueCount > 0) {
-            summary += `Identified ${issueCount} systemic issue(s). `;
-        }
-        if (contradictionCount === 0 && issueCount === 0) {
-            summary += `No significant issues detected.`;
-        }
-        return summary.trim();
-    }
-    /**
-     * Extract assumptions made during analysis
-     */
-    extractAssumptions(input) {
-        const assumptions = [
-            'All provided traces are authentic and unmodified',
-            'Timestamps are accurate and timezone-consistent (UTC)',
-            'Reported confidence values reflect agent certainty',
-        ];
-        if (!input.context?.historical_accuracy) {
-            assumptions.push('No historical accuracy baseline available - calibration assessment limited');
-        }
-        if (input.traces.length < 5) {
-            assumptions.push('Limited trace count may reduce statistical significance of findings');
-        }
-        if (!input.context?.correlation_groups?.length) {
-            assumptions.push('No correlation groups specified - assuming all traces are independent');
-        }
-        return assumptions;
-    }
-    /**
-     * Calculate confidence score based on analysis quality
-     */
-    calculateConfidence(output, input) {
-        let confidence = 0.7; // Base confidence
-        // Boost for sufficient trace count
-        if (input.traces.length >= 10) {
-            confidence += 0.1;
-        }
-        else if (input.traces.length >= 5) {
-            confidence += 0.05;
-        }
-        // Boost for multiple agents (cross-validation potential)
-        if (output.analysis_metadata.unique_agents >= 3) {
-            confidence += 0.05;
-        }
-        // Boost for having historical accuracy data
-        if (input.context?.historical_accuracy) {
-            confidence += 0.05;
-        }
-        // Reduce for limited scope
-        const scopeCount = Object.values(output.analysis_metadata.scope_executed)
-            .filter(Boolean).length;
-        if (scopeCount < 3) {
-            confidence -= 0.05;
-        }
-        return Math.min(1.0, Math.max(0.0, confidence));
-    }
-    /**
-     * Get constraints applied during analysis
-     */
-    getAppliedConstraints(input) {
-        const constraints = [
-            'read_only_analysis',
-            'no_output_override',
-            'no_correction_enforcement',
-            'no_logic_execution',
-            'deterministic_output',
-            'stateless_processing',
-        ];
-        // Add scope-specific constraints
-        if (input.scope.detect_contradictions) {
-            constraints.push('contradiction_detection_enabled');
-        }
-        if (input.scope.assess_confidence_calibration) {
-            constraints.push('calibration_assessment_enabled');
-        }
-        if (input.scope.identify_systemic_issues) {
-            constraints.push('systemic_analysis_enabled');
-        }
-        return constraints;
-    }
-    /**
-     * Classify error for proper error code
-     */
-    classifyError(error) {
-        if (error instanceof Error) {
-            if (error.name === 'ZodError') {
-                return contracts_1.AgentErrorCodes.VALIDATION_FAILED;
-            }
-            if (error.message.includes('persistence') || error.message.includes('ruvector')) {
-                return contracts_1.AgentErrorCodes.PERSISTENCE_ERROR;
-            }
-        }
-        return contracts_1.AgentErrorCodes.PROCESSING_ERROR;
-    }
-}
-exports.MetaReasonerAgent = MetaReasonerAgent;
-//# sourceMappingURL=meta-reasoner-agent.js.map