April 8, 2026 · API & AI
Claude MCP Stock Screener: StockMarketScan in Claude, Cursor and Continue
Set up a Claude MCP stock screener: connect Claude Desktop, Cursor or Continue to StockMarketScan and query screeners, patterns and options flow. 20 tools.
If you use Claude Desktop, Cursor, Continue, Windsurf, Zed or one of the other LLM clients that showed up over the last year, you know the routine. The model is helpful, but to get real data into the conversation you copy, paste, take screenshots or write yet another one-off script. For stock research, that friction is why most people never actually combine an LLM with a live screener or options flow tool.
The StockMarketScan MCP server is now live. Connect it to your LLM client and all 24 screeners, 25+ chart patterns, options flow data, unusual options activity, market momentum and the trend monitor become tools the model can call by itself, mid-conversation, whenever it needs fresh market data to answer you.
You don't need glue code, an agent framework or screener rows pasted into chat. You just ask.
Below: what MCP is, which tools a Claude MCP stock screener exposes, how authentication works, and the exact config for three clients.
What MCP is
The Model Context Protocol is an open standard, originally proposed by Anthropic, that lets any LLM client talk to external tools and data sources through one interface. Think of it as USB for AI tools. Claude Desktop, Cursor, Continue, Cline, Windsurf, Zed and dozens of other clients support it out of the box. Once a client is connected to an MCP server, the model sees a list of tools and decides, based on your question, which one to call and how.
For us, it's the simplest way to put StockMarketScan data in front of your model without tying you to one chat product. You keep the client you already use, and we show up as a tool source.
What a Claude MCP stock screener gives you: 20 tools in one server
The server exposes each core part of the platform as a named tool:
- Three screener tools list all 24 curated screeners, return the live rows of any one of them, and find stocks that appear in several screeners at once (which stocks are in both Golden Cross and Hot Prospects today?).
- Two chart pattern tools return every detected pattern on a symbol, or search a set of screeners for stocks forming a specific pattern (find me hot prospects with a cup and handle).
- Five options flow tools cover the daily overview with call/put ratios, consecutive-day streaks and premium, a historical timeline per symbol, a curated list filtered by screener overlap, and contract-level unusual options activity.
- Three stock info tools return fundamentals and quote data for any US symbol, plus raw OHLCV candles.
- Two composite tools combine the rest.
get_stock_reportcalls several endpoints in parallel and returns a full research brief.search_setupsranks the strongest bullish or bearish setups across all the data. - Three market context tools cover daily market momentum (advancers/decliners, new highs/lows), Trend Monitor output and ticker-to-ticker connections.
- One education tool explains platform terms like
strength_score,vol_oi_ratioandconsecutive_daysin plain language.
Add a ping tool for liveness checks and you get 20.
What you can do with it
A few examples that work today in any MCP-capable client:
"Which stocks appear in both hot-prospects and golden-cross today, and do any of them have a bullish chart pattern forming?"
The model calls search_stocks_in_screeners to find the overlap, passes those symbols to search_patterns to keep the bullish patterns, and summarizes the list.
"Show me AMD's options flow history for the last 30 days and tell me when the sentiment flipped."
The model calls get_options_flow_timeline for AMD with a 30-day window, reads the daily rows and tells you when the call/put ratio crossed above or below 1.0.
"Build a full stock report for NVDA: screeners it appears in, chart patterns, options flow state, and current market context."
The model calls get_stock_report, which hits six of our endpoints in parallel and returns a structured bundle. The model then writes it up in plain English.
"Which stocks are on today's curated bullish options flow list, and which ones have large-cap liquidity?"
The model calls get_options_flow_ranked for the curated list, looks up each symbol's fundamentals with get_stock_info, and filters by market cap.
Without MCP, each of these takes about ten browser tabs and a spreadsheet. The model does the coordination, we provide the data, and what you make of the answer is your call. Nothing the server returns is investment advice.
How authentication works: Bring Your Own Key
The rule is simple: you own your API key, we don't.
When you connect Claude Desktop or Cursor, the client sends your personal sms_* API key with every request as the X-API-Key header. The server checks it once per session and then forwards each tool call to api/v1/* for you. Rate limits, quota and tier are all tied to your key. You can rotate it on the Settings page at any time, and the server uses the new one on the next connection.
This is BYOK (Bring Your Own Key). We don't store a long-lived secret for your account, we can't run queries you didn't start, and nothing is shared between users. Each MCP session is isolated.
Installation: three clients, three configs
For Claude Desktop, edit claude_desktop_config.json (on macOS at ~/Library/Application Support/Claude/claude_desktop_config.json) and add this under mcpServers:
{
"mcpServers": {
"stockmarketscan": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.stockmarketscan.com/mcp"],
"headers": {
"X-API-Key": "sms_your_key_here"
}
}
}
}
Restart Claude Desktop. stockmarketscan now shows up in the tool picker, and the model can call any of the 20 tools mid-conversation.
In Cursor, go to Settings → Features → Model Context Protocol → Add New MCP Server and paste:
{
"name": "stockmarketscan",
"url": "https://mcp.stockmarketscan.com/mcp",
"headers": { "X-API-Key": "sms_your_key_here" }
}
For Continue (VS Code / JetBrains), edit ~/.continue/config.json and add an entry under experimental.modelContextProtocolServers with transport type sse, URL https://mcp.stockmarketscan.com/mcp, and the same header.
Full config snippets, including other clients, are in the install guide.
Which plan do I need
The MCP server works on Pro and Enterprise. Pro gives you 30 requests per minute and 2,000 per day across all screener, pattern, market and options flow tools. Enterprise raises that to 120 per minute and 20,000 per day. Basic is for personal use on the website and has no API or MCP access. Limits apply per API key, not per MCP session, so you can leave the client connected all day without changing your budget.
No account yet? Sign up and start a 7-day free trial. The MCP server needs a Pro or Enterprise key, but you can use the rest of the site to look around before subscribing.
What's next
This first version already covers every core part of the platform. Next on our list:
- Streaming prompts: prompt packs installed with the server, so you can type
/options-flow-report NVDAin Claude Desktop and get a structured answer. - More context tools: earnings calendar, economic calendar, sector performance, insider activity.
- Better caching. The server already caches the heavier endpoints in memory per session. We want a shared cache, so the first query for today's screener is cached for everyone.
- Multi-symbol tool calls.
get_stock_infocurrently runs once per symbol. Batch versions will cut the number of round trips the model has to make.
If you want something from the server, like a missing tool, a rate limit that doesn't fit how you work, or a client that's awkward to set up, tell us. MCP is a two-year-old protocol and the ecosystem is still settling. We'd rather build what heavy users actually need than guess.
Get started
1. Generate an API key at stockmarketscan.com/settings
2. Paste one of the configs above into your client
3. Ask your first question
If you get stuck, the full install guide with all client configs is at stockmarketscan.com/mcp.