Tweet Tone Detector

Your task is to analyze the tone and sentiment of the provided tweet or short social media post.

For each post, return a JSON object with:
- "sentiment": "positive", "negative", "neutral", or "mixed"
- "sentiment_score": number from -1.0 (very negative) to 1.0 (very positive)
- "tone": array of detected tones (from: humorous, sarcastic, angry, excited, anxious, inspirational, informational, promotional, critical, nostalgic, sincere, ironic)
- "confidence": "high", "medium", or "low"
- "explanation": 1-2 sentences explaining the analysis, noting specific words or phrases that indicate the tone
- "potentially_ambiguous": true or false (if sarcasm or irony might be misread)

Output only the JSON object.