- Home
- Skills
- Workflow Automation
- Shelly Home Automation
Works with the AI Agents you already use
Shelly Home Automation
Safe, repeatable Shelly control through the open-source shelly-mcp server: rooms, scenes, sunset schedules, energy costs.
$9
Shelly Home Automation
Example session with this skill installed
How much do my Shelly devices use, and what is it costing me? My tariff is 6.5 Kč per kWh.
- Read your context and instructions
- Compiled the shelly home automation
scripts/energy_report.py --config ~/.config/shelly-mcp/config.yaml --tariff 6.5 --currency Kč
Real run on five devices — Gen2 plugs and an RGBW strip, plus two Gen1 relays — read straight from the LAN:
televize switch:0 on 105.5 W 35.15 kWh 30d≈494 Kč lifetime 228 Kč
svetla-kuchyn switch:0 off 0.0 W 18.05 kWh 30d≈0 Kč lifetime 117 Kč
mycka meter:0 off 0.0 W 27.33 kWh 30d≈0 Kč lifetime 178 Kč
led rgbw:0 on 5.3 W 3.00 kWh 30d≈25 Kč lifetime 19 Kč
duo meter:0 off 0.0 W 0.00 kWh 30d≈0 Kč lifetime 0 Kč
Total live draw: 110.8 W ≈ 519 Kč per 30 days at this rate
Values match the MCP server's own shelly_energy_live (TV 35.147 kWh, dishwasher 27.331 kWh).
Suggestions
| Finding | Suggestion |
|---|---|
| The TV plug draws 105 W right now | If it stays on overnight: an on-device schedule 0 0 23 * * * → Switch.Set off (runs with the PC off) |
| LED strip in the evening | @sunset on / @sunset+2h off instead of a fixed time — follows the seasons |
| Dishwasher plug | Left out of any auto-off schedule — it runs long cycles overnight |
Nothing was switched — the report only reads status. Schedules are created after you confirm.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing a smart home via CLI or chat often feels like guesswork, leading to broken routines or devices switching off at the wrong time. This skill provides a safe, local-first interface for Shelly Gen1 through Gen4 devices by driving the shelly-mcp server with professional-grade caution.
What it does
- Device Control manages switches, dimmers, RGBW lighting, and motorized covers using absolute states for reliability.
- Scene Management bundles multiple device actions into named, deterministic routines like "Movie Night" that execute identically every time.
- On-Device Scheduling configures sunset, sunrise, and cron-based timers that run locally on the hardware without requiring a persistent server connection.
- Energy Analytics generates reports on live wattage, lifetime consumption, and cost projections using the bundled
energy_report.pyscript. - Safe Execution enforces manual confirmation for multi-room actions and prevents accidental power cycles on critical appliances like routers or freezers.
How it works
- Discovery runs
shelly_discoverto map the local network andshelly_list_devicesto pull friendly names from the configuration. - State Monitoring checks device status, firmware versions, and power draw before attempting any write operations.
- Command Execution uses the appropriate tool for the task, such as
shelly_switch_setfor toggles orshelly_schedule_createfor automations. - Validation reports per-device success or failure, diagnosing unreachable units or authentication errors immediately.
Frameworks & tools
This skill requires the shelly-mcp server installed via uvx. It interacts with Shelly Gen1, Gen2, Gen3, and Gen4 hardware over a local area network.
Why this beats prompting it yourself
Standard prompts often fail to distinguish between device generations or accidentally trigger unsafe "toggle" loops. This skill implements a strict safety playbook, uses hardware-native scheduling instead of fragile software timers, and handles local mDNS discovery that generic models cannot access.
Use cases
- Create a "Leaving Home" scene that shuts down all non-essential lighting and appliances.
- Schedule porch lights to follow seasonal sunset and sunrise times automatically.
- Audit household energy leaks by identifying devices with high standby power draw.
- Switch a device on for a fixed time — "TV for 10 minutes" — and let the device switch itself back off.
Known limitations
Schedules, scripts and webhooks need a local (LAN) connection; devices reachable only through Shelly Cloud get basic control and status. Gen1 devices cannot be scheduled through shelly-mcp. Sunset and sunrise schedules need shelly-mcp 0.1.3+ and the device location set.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 3 days ago
- Passed all security checks, Safe to install
Requires