Connect CannMenus to Claude
Set up CannMenus as a live data source in Claude. Ask about brands, pricing, and distribution in plain English.
What This Does
Connecting CannMenus to Claude gives you a direct line to real-time cannabis market data inside your Claude conversation. No dashboards, no exports - just ask questions in plain English and get answers pulled from live menu, pricing, and distribution data. CannMenus connects via the Model Context Protocol (MCP), an open standard that lets AI assistants use external tools and data sources.
MCP Server URL
This is the only URL you need. Copy it and follow the steps below for your platform.
How to Connect
Step 1
Open Connectors
Sign in to cannmenus.com first. Then, in Claude (claude.ai or the Claude desktop app), go to Customize > Connectors.
Step 2
Add a Custom Connector
Click the + button, then select Add custom connector. On a Claude Team or Enterprise plan, an Owner adds it once for everyone under Organization settings > Connectors > Add > Custom > Web.
Step 3
Enter the MCP Server URL
Type "CannMenus" as the connector name. Paste this URL: https://api.cannmenus.com/mcp/ — then click Add.
Step 4
Authorize with Your CannMenus Account
Click Connect. Claude opens a CannMenus page: sign in with your CannMenus account (the same one you use at cannmenus.com), then click Approve on the Authorize AI Access screen. Finish within about 10 minutes; if the session expires, click Connect again. This links your data permissions automatically via OAuth - no API key needed.
Step 5
Start a New Conversation
Open a new Claude conversation, click the + button at the bottom of the chat, choose Connectors, and make sure CannMenus is switched on. Ask "Use CannMenus whoami" to confirm which data your account can access.
Example Prompts to Try
Tips
- "What are the top 5 edible brands in California by retailer count?"
- "Which dispensaries in Colorado dropped Wyld products this month?"
- "Compare pricing for 1g vape carts in Michigan vs Illinois"
- "Show me brands gaining distribution in Florida"
- "What's the average flower price in Oregon right now?"
- "Run a gap analysis for Cookies in California - where do competitors have shelf space but Cookies doesn't?"
Which Model to Use
Tips
- Claude Opus: Best for deep, multi-step analysis sessions - market reports, competitive analysis, brand comparisons across states. Worth the slower speed for complex questions.
- Claude Sonnet: Great balance of speed and intelligence. Use for most day-to-day questions about brands, pricing, and distribution.
- Claude Haiku: Fastest option. Good for quick single-tool lookups like "What's the price of X in Y state?" but may struggle with multi-step analysis.
Getting the Best Results
Tips
- Always mention the state in your question. CannMenus data is organized by state, and queries without a state will be slower or fail.
- Check that CannMenus is active before asking. In Claude, click the + button at the bottom of the chat, choose Connectors, and switch CannMenus on.
- When Claude asks permission to use a CannMenus tool, click "Always allow" so you don't have to approve every data request in your session.
- Turn on extended thinking (the lightbulb icon) for complex questions that need multiple data pulls - like comparing brands across states or building a market report. It helps Claude plan which tools to call and in what order.
- Be specific about what you want: "Top 10 flower brands in Massachusetts by estimated sales, last 30 days" will get better results than "Tell me about Massachusetts."
- You can ask follow-up questions. Claude remembers the conversation context, so you can drill down: "Now break that down by subcategory" or "Which of those brands gained the most retailers?"
Available Tools
CannMenus provides 20+ tools covering product search, brand distribution, sales rankings, market trends, price comparisons, stock status, gap analysis, and company/MSO data. Claude will automatically pick the right tool based on your question. For sales and market share questions, it uses sales_rankings. For "where is this brand stocked?", it uses brand_stock_status. For "who carries X but not Y?", it uses brand_gap_analysis. You don't need to know the tool names - just ask your question naturally.