Kenki MCP Server

Model Context Protocol server that exposes the Kenki health platform to any MCP-compatible AI: Claude Desktop, OpenClaw, custom GPTs, Zed, Cline, etc.

What it does

Gives your AI assistant read + write access to your Kenki health data:

• Read all vitals, meds, sleep, food, water, measurements, appointments
• Read your profile and active goals
• Log water, food, vitals, sleep, measurements, supplements
• Create appointments
• Chat with Kenki's built-in AI (which has your full health context)

Setup

1. Create an agent token

• Go to https://kenki.ai/settings.html
• Scroll to Agent Tokens
• Click Create Token → select read:all and write:all scopes
• Copy the token (shown only once)

2. Install and configure

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (Mac) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "kenki": {
      "command": "npx",
      "args": ["-y", "kenki-mcp-server"],
      "env": {
        "KENKI_TOKEN": "your-token-from-step-1"
      }
    }
  }
}

Restart Claude Desktop. Kenki tools appear in the 🔨 menu.

OpenClaw

In OpenClaw's MCP settings, add:

• Command: npx -y kenki-mcp-server
• Env: KENKI_TOKEN=your-token

Local dev

export KENKI_TOKEN=your-token-here
export KENKI_URL=https://kenki.ai   # optional, default
node index.js

Available Tools

| Tool | Description | |------|-------------| | get_health_context | All recent health data | | get_user_profile | Name, units, goals | | list_actions | Discovery of what's loggable | | log_water | Log water intake | | log_food | Log a meal | | log_vitals | Log BP, HR, SpO2, temp, glucose | | log_sleep | Log a sleep session | | log_medication_taken | Record med adherence | | log_measurements | Weight, waist, arms, thighs | | create_appointment | Schedule an appointment | | add_supplement | Add to supplements list | | chat_with_kenki | Chat with Kenki's AI (uses your OpenAI/Claude/Gemini key) |

Example prompts to your AI

After setup, try asking your AI:

"Log that I drank 16oz of water" "What's my blood pressure trend over the past week?" "Schedule a dental cleaning for next Tuesday at 10am" "Log a grilled chicken salad with 450 calories for lunch" "What should I eat today based on my macro goals?" "Am I taking my medications consistently?"

Security

• Agent tokens are stored encrypted server-side (bcrypt hashed)
• All actions are audit-logged (agent_audit_log table)
• Scopes control what an agent can do (read:all, write:all, specific write scopes)
• Revoke tokens anytime from Settings > Agent Tokens
• Max 10 active tokens per user

Source

github.com/Kaisergxc/kenki