List 5 cat breeds. For each one, provide exactly: the breed name in bold, a one-sentence description, and a "cattitude score" from 1-10. Format each entry as a numbered list item. Do not include any text before or after the list.

Judge: Exactly 5 real breeds in a numbered list with bold names, one-sentence descriptions, numeric scores in range, and no surrounding text. Descriptions are accurate and distinctive, and the cattitude scores are thoughtfully calibrated to each breed's temperament (Ragdoll 2, Siamese 9). Fully compliant and engaging.

Judge: Exactly 5 real breeds with bold names, single-sentence descriptions, numeric scores (6, 8, 4, 7, 3) in range, and no text outside the list. Descriptions are accurate though slightly generic; formatting is fully compliant.

Judge: Exactly 5 real breeds in a numbered list, names bolded, each entry a single sentence that incorporates a numeric cattitude score, and no surrounding text. Descriptions are accurate and distinctive for each breed, though fairly conventional rather than especially vivid. A clean, fully compliant response.

Judge: Five real breeds, bolded names, numbered list, numeric scores 1-10, and no extra text before or after. Each entry is one sentence (the score is attached via a comma splice, slightly awkward grammatically but still a single sentence). Descriptions are accurate and more characterful than generic.

Judge: Exactly 5 real breeds as numbered list items, each with a bolded name, a single-sentence description, a numeric 1-10 cattitude score, and no text before or after the list. Descriptions are accurate and breed-distinctive, if a touch generic. All five hard constraints verified and satisfied.

Judge: Near-perfect format compliance: exactly 5 breeds, bold names, one-sentence descriptions, numeric cattitude scores (1-10), numbered list, no text before or after. Real breeds with accurate, distinctive descriptions. Clean execution.

Judge: Exactly 5 breeds, all real, names in bold, numbered list, each with a one-sentence description and cattitude score 1-10. No text before or after the list. Descriptions are accurate and capture each breed's character well. Perfect format compliance across all constraints.

Judge: Excellent format compliance: exactly 5 breeds, numbered list, bold breed names, each has one sentence description and a numeric cattitude score 1-10. No text before or after the list. All breeds are real with accurate, engaging descriptions. The cattitude scores are fun and opinionated (Siamese 8, Maine Coon 3, Persian 2 -- makes sense). Minor quibble: descriptions could be slightly more distinctive, but overall very clean execution.

Judge: Exactly 5 breeds, all bold, each with one-sentence description and cattitude score (1-10). No text before or after the list. Real breeds with accurate, engaging descriptions. All hard constraints met. The cattitude scores show personality (Siamese at 8 is right, Maine Coon at 3 tracks). Clean formatting throughout.

Judge: Exactly 5 breeds, numbered list, bold breed names, cattitude scores 1-10. No text before or after the list. However, each entry has a description that is arguably more than one sentence (some use commas that could be read as compound sentences). The descriptions are informative and breed-appropriate. Scores feel reasonable.

Judge: 5 real breeds, bold names, numbered list, cattitude scores 1-10. No text before or after. Descriptions are engaging and accurate. However, each entry has a description sentence AND a separate cattitude score line, and the descriptions are quite long (arguably more than 'one sentence' given compound structure). Minor format quibble but overall strong compliance.

Judge: Exactly 5 breeds, all real, bold names, numbered list, cattitude scores 1-10. Descriptions are engaging and capture each breed's character well. Minor issue: the prompt says 'a one-sentence description' but the descriptions include both a sentence AND a separate cattitude score line, which could be read as compliant. No text before or after the list. Scores feel well-calibrated (Siamese 9/10, Ragdoll 2/10).

Judge: 5 real breeds with bold names, numbered list, and cattitude scores 1-10. Descriptions are accurate and informative. However, each entry uses a semicolon to join the description and score, making it technically one long sentence plus a score fragment rather than 'exactly one sentence (description)' followed by a separate score. Also, the descriptions are slightly more than one sentence each. No extraneous text before or after the list.

Judge: Starts directly with the list (no preamble text -- good). 5 breeds, all real, bold names, each with a description and cattitude score. However, the cattitude scores are formatted as sub-bullets rather than inline with the description, and the descriptions run slightly longer than strictly 'one sentence.' Content quality is good -- breeds are distinct and descriptions are accurate.