Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@@jettyd/mcp Turn on the living room light"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for jettyd — gives AI agents direct access to IoT devices.
Connect Claude Desktop, Cursor, Continue, or any MCP-compatible client to your ESP32 devices.
npx @jettyd/mcpOr globally:
npm install -g @jettyd/mcpRelated MCP server: Nordic Thingy:52 MCP Server
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"jettyd": {
"command": "npx",
"args": ["@jettyd/mcp"],
"env": {
"JETTYD_API_KEY": "your-api-key-here"
}
}
}
}Add to your MCP config:
{
"jettyd": {
"command": "npx @jettyd/mcp",
"env": { "JETTYD_API_KEY": "your-api-key" }
}
}Tool | Description |
| List all devices with status |
| Get current sensor readings |
| Send command to device (relay, LED, etc.) |
| Historical sensor readings |
| Push JettyScript automation rules |
"What's the temperature in the greenhouse?" → Calls
read_device, reads shadow, returns current temperature
"Turn on the irrigation relay for 30 seconds" → Calls
send_commandwithrelay.onand{duration: 30000}
"Alert me if temperature goes above 30°C" → Calls
push_ruleswith a threshold rule
"Show me the humidity trend for the last 24 hours" → Calls
get_telemetrywith metric=air.humidity, period=24h
Sign up at jettyd.com and get your API key from the dashboard.
New to jettyd? The OpenClaw Blueprint walks through the complete setup in 6 steps — hardware, platform, OpenClaw connection, sensors, automations, and dashboard.
github.com/jettydiot/jettyd-mcp
MIT licence
Get historical sensor readings for a device over a time period.
Inputs: device_id (required), metric, period.
List all IoT devices in your jettyd account with their current status and last seen time.
Inputs: no named parameters.
Push JettyScript automation rules to a device. Rules run on-device and can trigger alerts when sensor thresholds are crossed.
Inputs: rules (required), device_id (required).
Get the current sensor readings and state for a specific device.
Inputs: device_id (required).
Send a command to a device (e.g. turn on a relay, blink an LED, set a value).
Inputs: action (required), device_id (required), params.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 5 tools
Each tool targets a distinct aspect of device interaction: listing devices, reading current state, sending commands, fetching historical data, and managing automation rules. No two tools have overlapping purposes.
All tool names follow a consistent verb_noun pattern (list_devices, read_device, send_command, get_telemetry, push_rules). Predictable and uniform.
5 tools is well-scoped for an IoT device management server, covering essential operations without redundancy or bloat.
The set covers listing, reading, commanding, telemetry, and rule automation, which are core for IoT interaction. Device create/update/delete are absent but likely outside the server's scope of managing existing devices.
One MCP server for the services your AI needs: memory, files, Microsoft 365, kanban and more.
MCP server unifying ERPs, CRMs, APIs and knowledge base for Claude, ChatGPT and Gemini.
MCP server for building and testing AI agents with multi-model experimentation and insights.
MCP server for AI dialogue using various LLM models via AceDataCloud