Homey

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill’s behavior matches its stated Homey smart-home control purpose, but it gives an agent real control over devices and uses Homey tokens.

Install this only if you want an agent to read and control your Homey smart-home devices. Be especially careful with locks, thermostats, security devices, and flows; require confirmation for those actions and keep Homey tokens private.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If the agent misunderstands a request or is given an unsafe instruction, it could change physical device states such as locks, lights, thermostats, or automations.

Why it was flagged

The skill can issue real smart-home control commands, including unlocking a door, which is high-impact but consistent with the documented Homey-control purpose.

Skill content
homeycli device "Front Door" set locked false
Recommendation

Use this skill only with trusted agents, require explicit confirmation for locks/security/thermostat changes, and prefer exact device IDs for high-impact actions.

What this means

Anyone with access to the stored token or config may be able to control or view Homey devices within that token’s permissions.

Why it was flagged

The skill uses Homey local/cloud tokens and may store them in a local config file; this is expected for the integration and disclosed in the docs.

Skill content
"local": { "address": "http://192.168.1.50", "token": "LOCAL_API_KEY" },
  "cloud": { "token": "CLOUD_TOKEN" }
Recommendation

Store tokens securely, avoid sharing ~/.homey/config.json, use the least-privileged token available, and revoke/rotate tokens if exposed.

What this means

Running release/developer scripts from an untrusted source could execute local shell commands.

Why it was flagged

A static scan found shell command execution in a release script. The provided user-facing setup and runtime commands do not show this script being automatically executed.

Skill content
return execSync(cmd, { stdio: 'pipe', encoding: 'utf8', ...opts }).trim();
Recommendation

Normal users should not run release scripts; review package scripts and source before running developer or publishing commands.