Connect your AI to live markets

Market data your AI agents can actually use

Connect Claude, ChatGPT, Cursor, Gemini, the OpenAI API, or any MCP-compatible agent to real-time and historical market data across seven asset classes — crypto, forex, stocks, indices, ETFs, commodities, and bonds. One config block, one API key or OAuth sign-in, zero invented endpoints.

  • 10 typed tools
  • 7 asset classes
  • Hosted — nothing to run
  • Included in every plan

Connect in under a minute

Point your client at the hosted endpoint with your API key — or, in OAuth-capable clients like the ChatGPT app, with no key at all: paste the URL, choose OAuth, and sign in with your TickerLayer account when prompted. That’s the whole integration — no SDK, no separate agent credentials.

MCPhttps://mcp.tickerlayer.com/mcp
claude mcp add --transport http tickerlayer https://mcp.tickerlayer.com/mcp \
  --header "x-api-key: YOUR_API_KEY"

How it works

1

Get a key

Sign up free and copy your API key from the dashboard. The same key works for REST, WebSocket, and MCP.

2

Paste the config

Add the tickerlayer server to Claude, Cursor, Gemini, the OpenAI API, or any MCP client. Ten tools register instantly.

3

Ask in plain language

“Compare EURUSD and GBPUSD this month.” The agent picks the right tool, symbol, and interval — and gets real data back.

One MCP, seven asset classes

Most providers make agents juggle a separate integration per asset class. TickerLayer serves everything through one consistent schema — the same symbols, timestamps, and response shapes everywhere.

  • CryptoBTCUSD · ETHUSD
  • ForexEURUSD · USDTRY
  • StocksUS:AAPL · SA:2222
  • IndicesUS500 · DE40
  • ETFsUS500ETF
  • CommoditiesXAUUSD · WTIUSD
  • BondsUS:10Y · DE:10Y
One schema, everywhere
GET /crypto/snapshot/BTCUSD
GET /stocks/snapshot/US:AAPL
GET /commodities/snapshot/XAUUSD

// same response shape, every asset class
{
  "symbol":         "BTCUSD",
  "bid":            64361.86,
  "ask":            64361.87,
  "last_price":     64361.86,
  "prev_close":     63230.00,
  "change_percent": 1.79
}

The tools your agent gets

get_quoteLatest bid/ask across crypto, forex, stocks, indices, ETFs, and commodities.
get_snapshotLast trade, quote, previous close, and change in one call.
get_last_tradeMost recent trade with price, size, and timestamp.
get_historyMinute, hour, and day OHLCV bars between two dates.
get_previous_closePrevious completed daily bar.
list_symbolsDiscover enabled symbols, with search.
get_market_statusOpen, closed, pre- and post-market by region.
get_market_sessionsSession windows for any market and date.
get_market_holidaysClosures and early closes by region.
get_bond_yieldGovernment bond yields by country and tenor.

Built agent-native, not bolted on

No hallucinated endpoints

Typed tools replace guessed HTTP paths. For codegen workflows, our llms-full.txt gives assistants the complete, real API surface.

Same key, same limits

MCP calls go through the exact entitlement and rate-limit path as REST. Your plan is the single source of truth — no shadow quota to manage.

Coverage the big APIs skip

Agents can pull markets Western providers often miss — including Saudi-listed stocks (SA:2222) and other MENA and Asia listings. Browse symbols.

Frequently asked questions

What is an MCP server for market data?

The Model Context Protocol (MCP) is the open standard AI assistants like Claude, ChatGPT, and Gemini use to call external tools. A market data MCP server exposes price, history, and market-calendar endpoints as typed tools, so an agent can fetch a live quote or historical bars directly instead of generating HTTP code and guessing endpoint paths.

Which AI clients can connect to the TickerLayer MCP server?

Any MCP client that supports custom HTTP headers: Claude Code, Cursor, Gemini CLI, VS Code with GitHub Copilot, Windsurf, Grok, and custom agents built on the Claude API, OpenAI Agents SDK, LangChain, or the Vercel AI SDK. Claude Desktop works through the documented local mcp-remote bridge. OpenAI models connect through the OpenAI Responses API's hosted MCP tool. And because the server now implements the MCP authorization spec (OAuth 2.1), the ChatGPT app connects natively as a developer-mode connector — as can any other OAuth-capable MCP client: paste the URL, sign in, no API key needed. The hosted endpoint speaks Streamable HTTP, the current MCP transport standard.

Can I use it inside the ChatGPT app?

Yes. The hosted server supports OAuth 2.1, which is what ChatGPT's developer-mode custom connectors expect. In ChatGPT on the web, open Settings → Apps & Connectors → Advanced settings and enable Developer mode (available on Plus, Pro, Business, Enterprise, and Edu). Choose Create, name the connector TickerLayer, set the MCP server URL to https://mcp.tickerlayer.com/mcp, pick OAuth as the authentication method, and create it — ChatGPT redirects to the TickerLayer sign-in, and you approve access once. In a chat, open the + menu → Developer mode and enable TickerLayer. Requests run under your TickerLayer account's plan, rate limits, and monthly quota.

Does MCP access cost extra?

No. The MCP server authenticates with your normal TickerLayer API key and is included with every plan, including the free tier. Tool calls pass through the same plan entitlements and rate limits as direct REST calls.

What asset classes can an AI agent query?

All of them: crypto, forex, global stocks (including Saudi Arabia and other MENA and Asia markets), indices, ETFs, commodities, and government bond yields — plus market status, trading sessions, and holiday calendars. One integration covers everything; most competing providers need one integration per asset class.

How do agents avoid hallucinating endpoints?

Two layers. Connected agents call typed MCP tools, so there is no path to invent. For code generation, we publish llms.txt and llms-full.txt — a complete, machine-readable reference of every real endpoint, symbol convention, and response shape — at tickerlayer.com/llms-full.txt.

Is the data suitable for AI trading agents?

TickerLayer provides derived, indicative, non-exchange-sourced market data for dashboards, analytics, research, and agent tooling. It is not an execution venue and does not provide official exchange feeds, so treat it as an information layer and validate suitability for your use case.

Give your agents market awareness today

Free tier included — connect an agent before you ever talk to sales.