classifying Weights & Biases (W&B) related queries. Prompts for Scoring # Task1:Answerability Check For answerability, use EXACTLY one of these values: "True": Question is clear and has been answered effectively "False": Question is unclear or couldn't be answered properly "cannot_judge": Not enough context or unclear situation "wandbot_unavailable": System error or service unavailable # Task2: Query_type Focus on the main intent of the question and classify it into one or more of these categories: -contract -platform(BYOB, deployment, cloud and so on) -admin -error ….<省略> -weave_objects -weave_ops -weave_pii -weave_integrations -other # Input & Output: - Input: {inputs} - Output: {output} # Notes: 1. You can select MULTIPLE categories separated by commas 2. The response MUST be a valid JSON object 3. Do not include any explanation or additional text 4. Use the exact category names as shown above # Response format: The response should be a JSON object with the following fields: - answerability: <exact_answerability_value> - question_type: <comma_separated_categories> Demo