Analyze Crypto Sentiment and Social Volume

This MCP server delivers cryptocurrency sentiment analysis to AI agents using Santiment's aggregated social media and news data, tracking market mood and surfacing emerging trends. Five tools cover this: get_sentiment_balance(asset, days=7) returns the average positive-versus-negative sentiment balance for an asset over a period; get_social_volume(asset, days=7) returns total social media mention counts; alert_social_shift(asset, threshold=50.0, days=7) detects significant spikes or drops in social volume compared to the recent average, flagging when a shift crosses the given percentage threshold; get_trending_words(days=7, top_n=5) returns the top trending terms in crypto discussions ranked by score; and get_social_dominance(asset, days=7) measures what percentage of crypto media discussion a given asset occupies. In practice, a natural-language question like "What's the sentiment balance for Bitcoin over the last week?" or "Tell me if there's been a big change in Bitcoin's social volume recently, with a 30% threshold" maps directly onto these tools and returns a plain-language answer (e.g. "Bitcoin's social volume spiked by 75.0% in the last 24 hours, from an average of 1,000 to 1,750").

Use this when an AI agent needs a read on crypto market mood or attention - sentiment trend for a specific asset, a spike/drop alert on social mentions, which terms are trending across crypto discussion, or how much of the conversation one asset dominates.

All data comes from Santiment, so results reflect Santiment's own aggregation methodology and coverage rather than a direct measurement of every social platform; sentiment and volume figures should be treated as a market-mood signal, not a substitute for on-chain or price data.

Sentiment balance, social volume tracking with configurable lookback windows, spike/drop alerting against a percentage threshold, top-N trending word extraction, and social dominance measurement - all queryable per-asset and per-time-period through the five tools above.

Requires Python 3.10+ and a Santiment API key (free or paid tier, from Santiment's own site). Clone the repository (git clone https://github.com/kukapay/crypto-sentiment-mcp.git), then configure your MCP client to run it with uv, pointing at the cloned directory and passing SANTIMENT_API_KEY as an environment variable. The project is MIT-licensed.

Traders, researchers, and crypto-focused AI agents that need social sentiment, volume, and trending-topic signals for specific assets rather than price data alone.

{
  "mcpServers": {
    "crypto-sentiment-mcp": {
      "command": "uv",
      "args": ["--directory", "path/to/crypto-sentiment-mcp", "run", "main.py"],
      "env": {
        "SANTIMENT_API_KEY": "your_api_key_here"
      }
    }
  }
}