How do I add a fund lookup tool to a Vercel AI SDK chatbot?

Route handler

typescript
// app/api/chat/route.ts — npm install ai @ai-sdk/openai @fundfactsapi/sdk zod
import { openai } from "@ai-sdk/openai";
import { streamText, tool } from "ai";
import { FundFacts } from "@fundfactsapi/sdk";
import { z } from "zod";
​
const ff = new FundFacts({ apiKey: process.env.FUNDFACTS_API_KEY! }); // server only
​
export async function POST(req: Request) {
  const { messages } = await req.json();
  const result = streamText({
    model: openai("gpt-4.1"),
    system: "You answer questions about investment funds using the tools. Always mention asOf with figures; say 'not disclosed' for empty fields.",
    messages,
    tools: {
      fundFacts: tool({
        description: "Current factsheet of a fund or ETF by 12-character ISIN.",
        inputSchema: z.object({ isin: z.string().length(12) }),
        execute: async ({ isin }) => {
          const f = await ff.getFund(isin); // cached in memory until expiresAt
          const d = f.data;
          return {
            name: f.name, category: d.profile.category, risk: d.riskRating, ter: d.headlineMetrics.ter, aum: d.keyFacts.aum,
            topHoldings: d.topHoldings.slice(0, 10), sectors: d.sector.slice(0, 8), countries: d.geography.slice(0, 8),
            calendarReturns: d.calendarReturns, asOf: d.dataAsOf,
          };
        },
      }),
      searchFunds: tool({
        description: "Find ISINs by fund name or issuer. Free.",
        inputSchema: z.object({ query: z.string().min(2) }),
        execute: async ({ query }) => (await ff.search(query)).results.slice(0, 10).map((r) => ({ isin: r.isin, name: r.name, issuer: r.issuer })),
      }),
    },
  });
  return result.toUIMessageStreamResponse();
}

The tool result can be rendered as a card on the client (holdings table, sector donut) with @fundfactsapi/widgets, the same React components that render this site's factsheets.

Or connect the MCP server

The AI SDK can consume MCP tools directly:

typescript
import { experimental_createMCPClient as createMCPClient } from "ai";
​
const mcp = await createMCPClient({
  transport: { type: "http", url: "https://fundfactsapi.com/api/mcp", headers: { Authorization: `Bearer ${process.env.FUNDFACTS_API_KEY}` } },
});
const tools = await mcp.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

• A cold ISIN can take minutes on first load; set maxDuration on the route accordingly or pre-warm the funds your users ask about.
• One request per fund found; the SDK client caches until expiresAt so repeated questions in a session cost nothing extra.
• Never call the API from the browser; the key belongs in the route handler.