market-sentiment

Files (3)

• SKILL.md
• sentiment_examples.md · references/
• sentiment_analyzer.py · scripts/

SKILL.md (first lines)

# Crypto Market Sentiment

## Overview

This skill enables aggregation of news from popular cryptocurrency RSS feeds, performs sentiment analysis on the articles, and computes a market sentiment score ranging from -1 (highly negative) to +1 (highly positive), along with evidence-based explanations.

## Workflow

Follow these steps to analyze crypto market sentiment:

1. **Select RSS Feeds**: Choose popular crypto RSS feeds (see references/rss_feeds.md for a curated list).
2. **Fetch News**: Retrieve recent articles from the selected feeds.
3. **Analyze Sentiment**: Classify each article's sentiment as positive (+1), negative (-1), or neutral (0) based on content keywords and context.
4. **Calculate Score**: Compute the average sentiment score across all articles.
5. **Generate Explanation**: Provide evidence from the news items supporting the score.

## Sentiment Classification Guidelines

- **Positive (+1)**: News about adoption, launches, partnerships, ETF approvals, price rallies, regulatory wins, or technological breakthroughs.
- **Negative (-1)**: News about hacks, crashes, regulatory crackdowns, liquidations, delays, or criticisms.
- **Neutral (0)**: Factual updates, mixed outcomes, or speculative content without clear bias.

## Output Format

The skill outputs:
- **Sentiment Score**: Numerical value between -1 and 1.
- **Explanation**: Breakdown by feed/source, key positive/negative drivers, and overall market implications.

## Resources

### scripts/
- `sentiment_analyzer.py`: Python script to fetch RSS feeds, parse articles, and compute sentiment score. Run with `python sentiment_analyzer.py` to get automated results.

### references/
- `rss_feeds.md`: List of popular crypto RSS feeds with URLs and descriptions.
- `sentiment_examples.md`: Examples of sentiment classification for common news types.

From kukapay/crypto-skills, MIT.

More from kukapay/crypto-skills