NHANES-MCP is one site: connect your AI first, then try it in the browser

README.md

@@ -13,11 +13,14 @@ pinned: false
 
 ### NHANES-MCP: AI-assisted analysis of public US health survey data
 
-<div class="grid md:grid-cols-2 gap-4"><a href="https://actogram-nhanes-mcp-demo.hf.space/#try"><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ask.svg" class="dark:hidden my-0!" style="margin:0" width="100%" alt="Ask NHANES-MCP in your browser. Example: is short sleep linked to death rates? Adults who reported 6 hours of sleep or less on weekdays had a 20% higher death rate during follow-up than adults who reported more, 95% confidence interval 11% to 29%, adjusted for age and sex. An association, not a cause. Free, no sign-up. Research prototype."><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ask-dark.svg" class="hidden dark:block my-0!" style="margin:0" width="100%" alt="Ask NHANES-MCP in your browser. Example: is short sleep linked to death rates? Adults who reported 6 hours of sleep or less on weekdays had a 20% higher death rate during follow-up than adults who reported more, 95% confidence interval 11% to 29%, adjusted for age and sex. An association, not a cause. Free, no sign-up. Research prototype."></a><a href="https://actogram-nhanes-mcp-demo.hf.space/#use"><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-connect.svg" class="dark:hidden my-0!" style="margin:0" width="100%" alt="Connect your AI assistant: ChatGPT, Claude, Mistral and others call NHANES-MCP, which calculates survey-weighted statistics on the US health survey. Free, one address."><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-connect-dark.svg" class="hidden dark:block my-0!" style="margin:0" width="100%" alt="Connect your AI assistant: ChatGPT, Claude, Mistral and others call NHANES-MCP, which calculates survey-weighted statistics on the US health survey. Free, one address."></a></div>
-<div class="hidden md:block mt-4"><a href="https://actogram-nhanes-mcp-demo.hf.space/#compare"><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ways.svg" class="dark:hidden my-0!" style="margin:0" width="100%" alt="Three ways to use NHANES-MCP. In your browser: free, no sign-up, short answers, 16 ready examples, 9 languages. In your AI assistant: free, all 16 tools with all options, literature search. Full research app, in early access: an AI assistant built in, guided research workflows, saved chats."><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ways-dark.svg" class="hidden dark:block my-0!" style="margin:0" width="100%" alt="Three ways to use NHANES-MCP. In your browser: free, no sign-up, short answers, 16 ready examples, 9 languages. In your AI assistant: free, all 16 tools with all options, literature search. Full research app, in early access: an AI assistant built in, guided research workflows, saved chats."></a></div>
-<div class="md:hidden mt-4"><a href="https://actogram-nhanes-mcp-demo.hf.space/#compare"><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ways-phone.svg" class="dark:hidden my-0!" style="margin:0" width="100%" alt="Three ways to use NHANES-MCP. In your browser: free, no sign-up, short answers, 16 ready examples, 9 languages. In your AI assistant: free, all 16 tools with all options, literature search. Full research app, in early access: an AI assistant built in, guided research workflows, saved chats."><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ways-phone-dark.svg" class="hidden dark:block my-0!" style="margin:0" width="100%" alt="Three ways to use NHANES-MCP. In your browser: free, no sign-up, short answers, 16 ready examples, 9 languages. In your AI assistant: free, all 16 tools with all options, literature search. Full research app, in early access: an AI assistant built in, guided research workflows, saved chats."></a></div>
+<div class="grid md:grid-cols-2 gap-4"><a href="https://actogram-nhanes-mcp-research.hf.space/#connect"><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-connect.svg" class="dark:hidden my-0!" style="margin:0" width="100%" alt="Connect your AI assistant: ChatGPT, Claude, Mistral and others call NHANES-MCP, which calculates survey-weighted statistics on the US health survey. Free, one address, one click in Claude or Cursor."><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-connect-dark.svg" class="hidden dark:block my-0!" style="margin:0" width="100%" alt="Connect your AI assistant: ChatGPT, Claude, Mistral and others call NHANES-MCP, which calculates survey-weighted statistics on the US health survey. Free, one address, one click in Claude or Cursor."></a><a href="https://actogram-nhanes-mcp-research.hf.space/"><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ask.svg" class="dark:hidden my-0!" style="margin:0" width="100%" alt="Ask NHANES-MCP in your browser. Example: is short sleep linked to death rates? Adults who reported 6 hours of sleep or less on weekdays had a 20% higher death rate during follow-up than adults who reported more, 95% confidence interval 11% to 29%, adjusted for age and sex. An association, not a cause. Try it: 5 questions without signing up. Research prototype."><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ask-dark.svg" class="hidden dark:block my-0!" style="margin:0" width="100%" alt="Ask NHANES-MCP in your browser. Example: is short sleep linked to death rates? Adults who reported 6 hours of sleep or less on weekdays had a 20% higher death rate during follow-up than adults who reported more, 95% confidence interval 11% to 29%, adjusted for age and sex. An association, not a cause. Try it: 5 questions without signing up. Research prototype."></a></div>
+<div class="hidden md:block mt-4"><a href="https://actogram-nhanes-mcp-research.hf.space/#ways"><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ways.svg" class="dark:hidden my-0!" style="margin:0" width="100%" alt="Three ways in. In your AI assistant: free over MCP, all 16 tools, one address. In your browser: 5 questions without signing up, every skill, 9 languages. Free account with a work e-mail, ready at once: 200 questions a day, saved chats, your own files."><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ways-dark.svg" class="hidden dark:block my-0!" style="margin:0" width="100%" alt="Three ways in. In your AI assistant: free over MCP, all 16 tools, one address. In your browser: 5 questions without signing up, every skill, 9 languages. Free account with a work e-mail, ready at once: 200 questions a day, saved chats, your own files."></a></div>
+<div class="md:hidden mt-4"><a href="https://actogram-nhanes-mcp-research.hf.space/#ways"><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ways-phone.svg" class="dark:hidden my-0!" style="margin:0" width="100%" alt="Three ways in. In your AI assistant: free over MCP, all 16 tools, one address. In your browser: 5 questions without signing up, every skill, 9 languages. Free account with a work e-mail, ready at once: 200 questions a day, saved chats, your own files."><img src="https://huggingface.co/spaces/Actogram/README/resolve/main/card-ways-phone-dark.svg" class="hidden dark:block my-0!" style="margin:0" width="100%" alt="Three ways in. In your AI assistant: free over MCP, all 16 tools, one address. In your browser: 5 questions without signing up, every skill, 9 languages. Free account with a work e-mail, ready at once: 200 questions a day, saved chats, your own files."></a></div>
 
-<sub>Example on the first card: US adults aged 20 and over (pregnant women excluded), NHANES 2005-2006 to 2013-2014, 27,798 participants and 4,073 deaths, median follow-up 9.2 years. Weekday sleep of 6 hours or less compared with more than 6 hours (the question NHANES asked until 2014); survey-weighted Cox model adjusted for age and sex; hazard ratio 1.20, 95% CI 1.11 to 1.29. An association in survey data, not a cause and not a prediction for any one person. Run: `cox_model({"exposure": ["Sleep hours, weekdays (2005-2014) <= 6"]})` on data snapshot data-v1 with the NCHS definitions (data af5de79980bb).</sub>
+**One click:** [Add to Claude](https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=NHANES-MCP&connectorUrl=https%3A%2F%2Factogram-nhanes-mcp.hf.space%2Fmcp) | [Add to Cursor](https://cursor.com/en/install-mcp?name=nhanes&config=eyJ1cmwiOiAiaHR0cHM6Ly9hY3RvZ3JhbS1uaGFuZXMtbWNwLmhmLnNwYWNlL21jcCIsICJ0eXBlIjogImh0dHAifQ%3D%3D) | [Install in VS Code](https://insiders.vscode.dev/redirect/mcp/install?name=nhanes&config=%7B%22type%22%3A%20%22http%22%2C%20%22url%22%3A%20%22https%3A%2F%2Factogram-nhanes-mcp.hf.space%2Fmcp%22%7D).
+Any other assistant takes the address `https://actogram-nhanes-mcp.hf.space/mcp`, with no sign-in. Questions or problems: admin@actogram.ai
+
+<sub>Example on the browser card: US adults aged 20 and over (pregnant women excluded), NHANES 2005-2006 to 2013-2014, 27 798 participants and 4 073 deaths, median follow-up 9.2 years. Weekday sleep of 6 hours or less compared with more than 6 hours (the question NHANES asked until 2014); survey-weighted Cox model adjusted for age and sex; hazard ratio 1.20, 95% CI 1.11 to 1.29. An association in survey data, not a cause and not a prediction for any one person. Run: `cox_model({"exposure": ["Sleep hours, weekdays (2005-2014) <= 6"]})` on data snapshot data-v1 with the NCHS definitions (data af5de79980bb).</sub>
 
 ### Activity research