mcpField
M
market-sentiment

market-sentiment

github.com/kukapayData & databases37623 installsUpdated 25w ago
Repository

About market-sentiment

A skill for estimating cryptocurrency market sentiment from recent articles in selected RSS feeds. It classifies articles, averages their scores from -1 to +1, and explains the result using evidence from the news.

Key features

  1. 01Selects cryptocurrency RSS feeds
  2. 02Fetches recent articles
  3. 03Classifies positive, negative, and neutral news
  4. 04Calculates an average sentiment score
  5. 05Explains scores with news evidence

Use cases

  1. 01Summarizing sentiment across selected feeds
  2. 02Reviewing positive and negative news drivers
  3. 03Producing an evidence-based market sentiment estimate

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/kukapay/crypto-skills --skill market-sentiment

After installing, type/market-sentimentin Claude Code to confirm it loaded.

Bundled filesreferences/sentiment_examples.mdscripts/sentiment_analyzer.py

Details

From the project README

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.

Comments

Markdown supported