Connect your AI to live earnings-volatility data
The EarningsWatcher MCP server gives ChatGPT, Claude, and Cursor direct access to the same data the platform runs on: live implied moves, IV rush and IV crush metrics, 10 years of implied-vs-actual earnings history, the scanner, the simulator, and the backtester — 90+ tools, no API key, one URL.
Included with membership
MCP access is a member feature. Every EarningsWatcher plan includes it — once you're in, connecting your AI takes about two minutes and the step-by-step instructions are waiting in your member dashboard (works with ChatGPT, Claude, Claude Code and Cursor).
Become a member & connect your AI →
Want to see it working first? Read how one member runs his entire trading workflow through it: MCP case study: a member’s trading OS — or jump to the summary below.
What you can ask
- "Which names are on the IV rush radar this week?"
- "How much does JPM usually move on earnings, and what's implied right now?"
- "Run the scanner for long-volatility setups reporting this week."
- "Simulate a straddle on next week's biggest implied move."
- "Backtest short-volatility entries on names that rarely beat their implied move."
- "Add the top IV-rush name to my watchlist and journal the idea."
Why traders use it
- Research where you already work. Your AI assistant becomes an earnings research terminal — no tab-switching, no copy-pasting screenshots.
- Real data, not model guesses. Answers come from EarningsWatcher's live pipeline (options-market implied moves, historical realized moves, beat rates) instead of the model's memory.
- Full workflow. Read the data, then act on it: watchlist, planner, paper trades, journal — the same member workflow as the app.
See it in practice: a member’s trading OS
littlebeard, lifetime member, runs his whole process from the chat. He connected EarningsWatcher to his AI once, turned what he had learned into heuristics his assistant can search, and then ran a real IV Rush paper trade end to end — the research, the entry, the exit, the journal entry — without opening a dashboard.
It is not an “ask ChatGPT for picks” story. It shows what the 90+ tools look like when they are wired into a routine: the radar in the morning, the simulator before a ticket, the journal after the close, all in one conversation.
Read the case study: how a member turned the MCP into a personal trading OS →Two trades planned in chat, then checked against the data. In July 2026 another member asked the assistant for an earnings trade twice: a long straddle on NFLX and a short strangle on SOFI, both paper-traded from the conversation. We went back to our own implied-move history to see how each ended. One won at the planned exit, one lost, and the loss is the more useful lesson: the thesis was built on peak moves and cashed at the close.
Read it: two earnings trades planned with an AI assistant, and what happened next →Frequently asked questions
What is the EarningsWatcher MCP server?
It is a Model Context Protocol (MCP) server that lets AI assistants like ChatGPT, Claude, and Cursor query EarningsWatcher's live earnings-volatility data directly: implied moves, IV rush and IV crush metrics, historical earnings moves, beat rates, the scanner, the simulator, and backtests.
Which AI apps work with it?
Any MCP-capable client works, including ChatGPT (Connectors), Claude and Claude Code, and Cursor. MCP access comes with membership; connection instructions are in the member dashboard.
Do I need an API key?
No API key is needed. You sign in with your normal EarningsWatcher email and password. MCP access is included with an EarningsWatcher membership.
What can I ask once connected?
Examples: which names are on the IV rush radar this week, how much a stock usually moves on earnings, its current implied move, long-volatility scanner results, a simulated earnings position, or a backtest of an earnings strategy.
Is this an MCP server for stock market data?
Yes — EarningsWatcher is an MCP server specialized in stock market earnings data: implied and expected moves, IV rush and IV crush, 10 years of implied-vs-actual move history, an earnings calendar, a scanner and a backtester. It is built for trading around earnings rather than general quotes.
Give your AI an earnings edge
Membership includes the platform, the MCP server, and the full data pipeline behind both.