How do I add fund and ETF data as a tool in LangChain or LangGraph?

A tool in twenty lines

python
# pip install git+https://github.com/fundfactsapi/fundfactsapi-python langchain langchain-openai
import os
from langchain_core.tools import tool
from fundfacts import FundFacts
​
ff = FundFacts()  # reads FUNDFACTS_API_KEY
​
@tool
def fund_facts(isin: str) -> dict:
    """Current factsheet of a fund or ETF by 12-character ISIN: name, TER, 1-7 risk
    indicator, category, fund size, top holdings, sector and country split,
    calendar and annualised returns, as-of date. Returns {"error": ...} when the
    ISIN is not a covered fund."""
    try:
        f = ff.get_fund(isin)
    except Exception as e:  # FundFactsError carries status and code
        return {"error": str(e)}
    d = f["data"]
    return {
        "name": f["name"], "category": d["profile"]["category"], "risk_1_to_7": d["riskRating"],
        "ter": d["headlineMetrics"]["ter"], "aum": d["keyFacts"]["aum"],
        "top_holdings": d["topHoldings"][:10], "sectors": d["sector"][:8], "countries": d["geography"][:8],
        "calendar_returns": d["calendarReturns"], "annualised_returns": d["annualisedReturns"],
        "as_of": d["dataAsOf"],
    }
​
@tool
def search_funds(query: str) -> list[dict]:
    """Find fund ISINs by name or issuer, e.g. 'msci world' or 'vanguard all-world'. Free."""
    return [{"isin": r["isin"], "name": r["name"], "issuer": r["issuer"]} for r in ff.search(query)["results"][:10]]

Bind the tools to a model and run an agent:

python
from langchain_openai import ChatOpenAI
from langgraph.prebuilt import create_react_agent
​
agent = create_react_agent(ChatOpenAI(model="gpt-4.1"), [fund_facts, search_funds],
    prompt="You answer questions about investment funds. Always quote as_of with figures. Say 'not disclosed' for empty fields; never estimate.")
result = agent.invoke({"messages": [("user", "What does IE00B4L5Y983 hold and what does it cost?")]})
print(result["messages"][-1].content)

Trimming the payload in the tool (the full one is a few kilobytes with the series) keeps the context small; return the whole data object if the agent needs the monthly series for calculations.

Or use the MCP server directly

python
# pip install langchain-mcp-adapters
from langchain_mcp_adapters.client import MultiServerMCPClient
​
client = MultiServerMCPClient({
    "fundfacts": {
        "transport": "streamable_http",
        "url": "https://fundfactsapi.com/api/mcp",
        "headers": {"Authorization": f"Bearer {os.environ['FUNDFACTS_API_KEY']}"},
    }
})
tools = await client.get_tools()  # get_fund, search_funds, compare_funds, analyze_portfolio, fund_overlap, get_scpi

The MCP server exposes seven tools, each counted like the REST endpoint it wraps: get_fund (ISIN → the full factsheet payload), search_funds (name → ISINs, free), compare_funds (side-by-side table for several ISINs), analyze_portfolio (weighted look-through of a list of positions, Starter and above), fund_overlap (shared holdings between funds, Starter and above), get_scpi (French SCPI data by name or ISIN) and search_scpi (SCPI name → matches, free), plus a tour prompt that explains the service on the example fund IE00B4L5Y983, which is not counted. Every tool only reads data. Authentication is either your FundFacts account (OAuth 2.1 with PKCE, the sign-in the claude.ai, Claude Desktop and ChatGPT connectors run) or the same bearer API key as the REST API, sent as an Authorization header by editors, CLIs and frameworks. The MCP server guide has the full tool reference, setup for every client and troubleshooting.

Notes for production agents

• The first call for a cold ISIN can take up to a few minutes; give the tool a 300-second timeout and tell the model a first lookup can be slow.
• One request per fund found. Cache tool results by ISIN until expiresAt if users repeat questions.
• The fund data for AI agents guide covers prompt design and guardrails in more depth.