@aiorg/mcp-financial-sentiment

MCP server for financial news and sentiment analysis. Provides tools for monitoring stock sentiment from news (Finnhub) and social media (StockGeist).

Features

• News Analysis: Fetch and analyze company news from Finnhub
• Social Sentiment: Get sentiment from Reddit and Twitter via StockGeist
• Narrative Detection: Search for specific themes/keywords in headlines
• Trigger Checking: Automatically evaluate sentiment-based investment triggers
• Caching: 15-minute cache to respect API rate limits

Installation

npx @aiorg/mcp-financial-sentiment

Or add to your Claude Code MCP configuration:

{
  "mcpServers": {
    "financial-sentiment": {
      "command": "npx",
      "args": ["-y", "@aiorg/mcp-financial-sentiment"],
      "env": {
        "FINNHUB_API_KEY": "${FINNHUB_API_KEY}",
        "STOCKGEIST_API_KEY": "${STOCKGEIST_API_KEY}"
      }
    }
  }
}

Setup

1. Finnhub API Key (FREE)

1. Go to finnhub.io/register
2. Create free account
3. Copy API key
4. Add to .env: FINNHUB_API_KEY=your_key_here

Free tier includes:

• 60 API calls/minute
• 1 year of historical news
• Real-time news updates

2. StockGeist API Key (FREE credits)

1. Go to api.stockgeist.ai/signup
2. Create account (10k free credits)
3. Copy API key
4. Add to .env: STOCKGEIST_API_KEY=your_key_here

Tools

get_news

Fetch recent news for a stock ticker.

Input: { ticker: "AMZN", days: 7 }
Output: {
  ticker: "AMZN",
  news: [
    { headline: "...", summary: "...", source: "...", datetime: "...", url: "..." }
  ],
  totalCount: 15
}

get_social_sentiment

Get social media sentiment from Reddit and Twitter.

Input: { ticker: "AMZN" }
Output: {
  ticker: "AMZN",
  sentiment: {
    score: -0.35,
    label: "bearish",
    volume: 1250,
    sources: { reddit: 800, twitter: 450 }
  }
}

check_narrative

Search for specific narratives/keywords in news headlines.

Input: {
  ticker: "AMZN",
  keywords: ["AWS lagging", "cloud share loss", "Google Cloud gains"],
  days: 7
}
Output: {
  matches: [
    { headline: "...", matchedKeywords: ["cloud share loss"], sentiment: "negative", url: "..." }
  ],
  narrativeStrength: "emerging"
}

check_triggers

Evaluate sentiment-based investment triggers.

Input: {
  ticker: "AMZN",
  triggers: [
    "Bearish social sentiment on AWS",
    "AWS lagging narrative in headlines"
  ]
}
Output: {
  ticker: "AMZN",
  triggersChecked: 2,
  triggersFired: 1,
  results: [
    { trigger: "...", type: "sentiment", status: "fired", details: "..." }
  ],
  recommendation: "1 trigger(s) fired! Review AMZN for potential action."
}

Use Cases

Contrarian Investing

Monitor when sentiment becomes overly negative for quality companies:

Use check_triggers with:
- "Bearish analyst sentiment"
- "Negative social media sentiment"
- "Share loss narrative dominates headlines"

Morning Briefing

Integrate with Investor OS /brief command to automatically check watchlist sentiment.

Earnings Watch

Before earnings, check narrative sentiment to gauge market expectations.

Caching

All API responses are cached for 15 minutes to:

• Respect API rate limits
• Speed up repeated queries
• Reduce costs on paid tiers

License

MIT