artifacts/api-server/src/lib/guide.ts

/**
 * guide.ts — best-of-N sample reranker.
 */

export interface Sample {
  text: string;
  score: number;
}

const STOP = new Set([
  'the','a','an','is','are','was','were','of','in','to','and','or','it','its',
  'this','that','with','for','on','as','by','be','at','from','but','not','if',
]);

export function rerankSamples(
  samples: Sample[],
  context: string,
  query: string,
): { best: Sample | null; bestScore: number } {
  if (!samples.length) return { best: null, bestScore: 0 };

  const terms = query.toLowerCase().match(/[a-z]{3,}/g) ?? [];
  const keywords = terms.filter(t => !STOP.has(t));

  let best = samples[0];
  let bestScore = -Infinity;

  for (const s of samples) {
    let score = s.score;
    const lower = s.text.toLowerCase();
    for (const kw of keywords) {
      if (lower.includes(kw)) score += 0.3;
    }
    // Penalise very short outputs
    const words = s.text.split(/\s+/).filter(Boolean).length;
    if (words < 5) score -= 2;
    if (score > bestScore) {
      bestScore = score;
      best = s;
    }
  }

  return { best, bestScore };
}