Auto-evolución

Documentación en español de Auto-evolución, alineada con el monorepo y la guía inglesa.

Entrada rápida

Execution completes
  │
  ├─ TrajectoryAnalyzer       ← Analyze execution patterns
  │   └─ Extract features: duration, tokens, success rate
  │
  ├─ MetaLearner              ← Thompson Sampling
  │   └─ Update Beta distributions per arm
  │
  └─ EvolverAgent             ← Cross-run optimization
      └─ Propose configuration changes

interface Arm {
  name: string;             // e.g., "topology:DISPATCH"
  alpha: number;            // Success count
  beta: number;             // Failure count
}

class MetaLearner {
  selectArm(arms: Arm[]): Arm {
    // Sample from each arm's Beta distribution
    // Pick the arm with the highest sampled value
    return arms.reduce((best, arm) => {
      const sample = BetaDistribution.sample(arm.alpha, arm.beta);
      return sample > best.sample ? { ...arm, sample } : best;
    });
  }

  updateArm(arm: Arm, success: boolean): void {
    if (success) arm.alpha++;
    else arm.beta++;
  }
}

const analyzer = new TrajectoryAnalyzer();
const patterns = analyzer.analyze(trajectory);

// patterns.bottlenecks     → [step names sorted by duration]
// patterns.commonFailures  → [failure modes by frequency]
// patterns.providerRanking → [providers sorted by success rate]

Metric	Impact
Topology selection accuracy	+15-20% after 1000 runs
Provider availability	Failover optimized for current latency
Verification threshold tuning	False positives reduced by 30%
Token efficiency	Budget allocation converges to task need

Notas

• CLI monorepo: cliEntry.ts · tras build: commander
• Métricas: 25 proveedores · 5 topologías · 18 tools · 6700+ tests
• Firmas API exactas: monorepo / API overview

Relacionado

• Arquitectura
• Inicio rápido