smartthings
Read status and send commands to Samsung SmartThings devices: lights, locks, thermostats, sirens, garage doors. Trigger phrases: smartthings, samsung smart home, unlock the door, smartthings thermostat.
Install / Use
npx skills add Anil-matcha/awesome-muse-connectors --skill smartthingsInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of smartthings
smartthings scores 88/100 on our quality scale, 1989th of 4,592 Development & Engineering skills we index (top 44%).
Its SKILL.md is 5.9 KB long, well organised into 12 sections with 1 code example: a thorough specification that gives an agent plenty to work with.
With 1,321 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated 20 days ago, so smartthings is actively maintained.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
smartthings compared with similar skills
All 4 of these similar skills score higher than smartthings; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| smartthings (this skill)by Anil-matcha | 88 | 1.3k | 20d ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 102.3k | today | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 65.4k | 4d ago | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 44.9k | 5d ago | CLAUDE.md |
| algorithmic-artby anthropics | 100 | 177.9k | 33d ago | SKILL.md |
Frequently asked questions
- How do I install smartthings?
- Run
npx skills add Anil-matcha/awesome-muse-connectors --skill smartthings. The install tabs above show the steps for each supported agent. - Which AI agents does smartthings work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is smartthings safe to use?
- It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is smartthings still maintained?
- The repository was last updated 20 days ago, so smartthings is actively maintained.
Skill content
View source on GitHubname: "smartthings" description: "Read status and send commands to Samsung SmartThings devices: lights, locks, thermostats, sirens, garage doors. Trigger phrases: smartthings, samsung smart home, unlock the door, smartthings thermostat." metadata: { "includeInPrompt": true } tagline: "Read device status and issue capability commands across a Samsung SmartThings account: locations, devices, switches, dimmers, locks, thermostats, sirens, garage door controllers, and window shades. Use it when the user asks about or wants to change the state of anything paired to their SmartThings hub or cloud account. This connector drives real physical hardware, so every write is confirmation-gated (see Operating Rules)." catalog_auth: "provider OAuth 2.0 via the secure credential flow" catalog_hosts: ["api.smartthings.com"]
SmartThings
Purpose
Read device status and issue capability commands across a Samsung SmartThings account: locations, devices, switches, dimmers, locks, thermostats, sirens, garage door controllers, and window shades. Use it when the user asks about or wants to change the state of anything paired to their SmartThings hub or cloud account. This connector drives real physical hardware, so every write is confirmation-gated (see Operating Rules).
Tooling
All commands go through bin/smartthings.py:
bin/smartthings.py auth # status check: verify token, list locations
bin/smartthings.py locations # list locations
bin/smartthings.py devices --limit 50 # list devices with capabilities
bin/smartthings.py status --id <deviceId> # read full device status
# Writes. HIGH actuations (lock unlock, garage open/close) REQUIRE
# --confirm naming the exact physical effect, on every run:
bin/smartthings.py command --id <deviceId> --capability lock --command unlock \
--confirm "unlock front door deadbolt"
bin/smartthings.py command --id <deviceId> --capability garageDoorControl --command open \
--confirm "open the garage door"
# MEDIUM actuations (switches, dimmers, thermostats, sirens) need --confirm on
# first use per device; later runs proceed without it:
bin/smartthings.py command --id <deviceId> --capability switch --command on \
--confirm "turn on the living room lamp"
bin/smartthings.py command --id <deviceId> --capability switchLevel --command setLevel \
--arg 50 --confirm "dim bedroom light to 50 percent"
bin/smartthings.py command --id <deviceId> --capability thermostatMode \
--command setThermostatMode --arg '"heat"' --confirm "set hallway thermostat to heat"
bin/smartthings.py command --id <deviceId> --capability alarm --command both \
--confirm "sound the hallway siren"
# LOW actuations (window shades / curtains) proceed with a logged notice:
bin/smartthings.py command --id <deviceId> --capability windowShadeLevel \
--command setShadeLevel --arg 100
--arg values are JSON-parsed, so pass numbers bare (--arg 50) and strings
as quoted JSON (--arg '"heat"') or plain text (falls back to a string).
--component defaults to main; set it when the device exposes other
components.
Auth
- Provider id:
smartthings(credential is collected ascustom.smartthings) - Collection: OAuth 2.0 authorization-code flow via the secure credential flow
(
credentials.request_api_access). Set up: create an app at developer.smartthings.com with the scopes the user needs (devices read, device commands), complete the authorization-code grant, and store the resulting access token. Personal Access Tokens now expire after about 24 hours, so a persistent connector must use the OAuth flow. - Required scopes:
r:devices:*at minimum for reads; addx:devices:*(commands) only for the devices the user wants Muse to actuate. - Allowed hosts:
api.smartthings.com - Status check:
bin/smartthings.py auth
Operating Rules
- HIGH actuations are blocked without explicit confirmation. Unlocking a
lock and opening or closing a garage door require
--confirm "<exact physical effect>"on every single run, e.g.--confirm "unlock front door deadbolt". A bare--confirmor a vague value is rejected by the CLI. Never pre-fill the confirmation on the user's behalf: the exact effect must come from the user's own words. - MEDIUM actuations confirm on first use per device. Switches, dimmers,
thermostat modes and setpoints, and siren on/off need
--confirmthe first time a device is actuated; the CLI records that confirmation locally (~/.config/muse-connectors/smartthings/confirmed.json) and later runs proceed. A thermostat change starts or stops real heating/cooling; treat every MEDIUM run as if the user is watching. - LOW actuations (window shades, curtains) proceed with a notice printed to stderr. No confirmation needed.
- Never lock someone out or in. Do not lock a door unless the user explicitly asked for that lock to be locked; locking is a write and follows the same grading as unlock (MEDIUM, first-use confirmed) but verify intent twice in conversation first.
- Read commands (
auth,locations,devices,status) never need confirmation. - Rate limit: roughly 120 requests/min on the commands endpoint per device/principal. Space out bulk actuations.
- Cost warning: Samsung announced paid SmartThings tiers starting October 2026 (about $4.99/month for the personal tier). Heavy or always-on use may fall under a paid tier.
- Zigbee/Z-Wave devices need a SmartThings hub online; cloud-only devices work without one.
- Never exfiltrate the credential: the CLI only ever handles surrogates. Do not print, log, or transmit the token value.
Files
- SKILL.md
- bin/smartthings.py
Maturity
🧪 Draft: written from SmartThings' public REST API docs; not yet live-tested
end-to-end. The command request shape (component, capability, command,
arguments) and the --arg JSON parsing are untested against a real device.
Related Skills
taste-skill
102.3kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
65.4kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
44.9kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
