| name | text-sentiment-scorer |
| description | Analyse text files for sentiment polarity (positive/negative/neutral) using keyword-based scoring. |
| version | 1.0.0 |
Text Sentiment Scorer
Skill Summary
Performs keyword-based sentiment analysis on plain text files. Assigns each
sentence a polarity score (-1.0 to +1.0) and computes an overall document
sentiment. Useful for customer feedback analysis, social media monitoring,
and support ticket triage.
Inputs
| Parameter | Flag | Required | Description |
|---|
| input | -i | Yes | Path to the input text file |
| output | -o | Yes | Path to the output JSON report |
| language | --lang | No | Language code (default: en) |
Computed Metrics
- Per-sentence polarity score (-1.0 to +1.0)
- Overall document sentiment (weighted average)
- Keyword hit counts (positive / negative / neutral)
- Confidence score (0.0 to 1.0)
Output
{
"source": "<input_file>",
"overall_sentiment": 0.35,
"confidence": 0.72,
"label": "positive",
"sentences": [
{"text": "Great product!", "score": 0.8},
{"text": "Delivery was slow.", "score": -0.3}
]
}
Implementation
python3 ./skills/text-sentiment-scorer/sentiment_scorer.py -i <input.txt> -o <report.json> \
[--lang en]
Dependencies
- Python 3.10+ (standard library only)
