DEV Community

Robin Beraud-Sudreau
Robin Beraud-Sudreau

Posted on

Build Financial Analysis into Your Hermes AI Agent with the FMP Skill

#hermes #ai #finance #opensource

Build Financial Analysis into Your Hermes AI Agent with the FMP Skill

The Problem

Building AI agents that can answer real-time financial questions is tough. You need reliable data sources, API integrations, and the right abstractions to make it seamless. That's why the Hermes community just shipped something powerful: the Financial Modeling Prep (FMP) skill.

Whether you're building a financial chatbot, an investment research assistant, or an autonomous trading dashboard, you now have a plug-and-play skill that gives your Hermes AI agent instant access to professional-grade financial data.

What Can You Do?

The FMP skill connects Hermes to real-time market data across multiple asset classes:

  • Stock quotes & fundamentals — real-time prices, company profiles, income statements
  • Financial ratios — P/E ratios, dividend yields, debt-to-equity metrics
  • Stock screeners — filter stocks by market cap, PE ratio, or custom criteria
  • Crypto & forex prices — Bitcoin, Ethereum, EUR/USD, and more
  • Market indices — CAC40, S&P 500, Nasdaq snapshots
  • News & movers — trending stocks and market sentiment

How It Works

The skill is built on a clean abstraction over Financial Modeling Prep's stable API (v2025+):

def fmp(path, **params):
    params["apikey"] = os.environ["FMP_API_KEY"]
    qs = "&".join(f"{k}={v}" for k, v in params.items())
    url = f"https://financialmodelingprep.com/stable/{path}?{qs}"
    return json.loads(urllib.request.urlopen(url).read())
Enter fullscreen mode Exit fullscreen mode

With this helper, querying data is one line:

# Get Apple's current stock price
price = fmp("quote", symbol="AAPL")
print(f"AAPL: ${price[0]['price']} ({price[0]['changesPercentage']:+.2f}%)")
Enter fullscreen mode Exit fullscreen mode

Using It with Hermes

Setup is straightforward:

  1. Get an API key from financialmodelingprep.com
  2. Add it to your Hermes config (~/.hermes/.env): 
   FMP_API_KEY=your_key_here
Enter fullscreen mode Exit fullscreen mode
  1. Ask your agent — now you can ask natural questions like:
    • "What's the current price of Tesla?"
    • "Show me the PE ratio for Microsoft"
    • "Find stocks with market cap > $100B and PE < 15"
    • "What's Bitcoin trading at right now?"

Hermes will automatically route these queries to the FMP skill and deliver formatted, contextual answers.

Why This Matters

Financial data is increasingly important for AI agents. This skill removes the friction of integrating FMP—you get:

  • ✅ Standardized endpoints — no guessing which API version to use
  • ✅ Error handling baked in — reliable queries that don't break
  • ✅ Community-maintained — open-source means transparency and improvements
  • ✅ Production-ready — using stable API endpoints

Get Started

The FMP skill is live in the hermes-skills repository. If you're building financial AI agents, check it out and give the repo a ⭐ to support the project.

Have ideas for other financial endpoints? PRs welcome—let's build the financial data layer for AI agents together.

Top comments (0)