Starling Home Hub (Nest/Google Home)
WarnAudited by ClawScan on May 10, 2026.
Overview
The skill is transparent and purpose-aligned, but it gives an agent powerful control over locks, cameras, thermostats, and home/away state without artifact-level confirmation safeguards.
Before installing, decide whether you want an agent to control physical smart-home devices. Use a read-only Starling API key unless you explicitly need writes, require confirmation for unlocking doors, changing cameras, thermostat settings, or home/away mode, keep the hub off the public internet, and protect any saved snapshots and API keys.
Findings (4)
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.
If invoked incorrectly or too broadly, the agent could change physical home state, including unlocking a door, changing heating/cooling, or modifying camera behavior.
The wrapper accepts arbitrary key=value property updates for any device ID and posts them to the hub. The API reference shows writable properties include door lock state, camera settings, thermostat controls, and home/away state, but the script has no separate confirmation or guardrail for these high-impact actions.
set <id> <key=value>... Set device properties ... api_post "/devices/${DEV_ID}" "$JSON" | fmt_jsonUse read-only Starling API keys by default, create separate narrowly scoped write keys only when needed, and require explicit user confirmation before any write action, especially lock, camera, thermostat, or home/away changes.
Anyone or any process with the API key could potentially read device state, access camera snapshots/streams, or change writable smart-home properties allowed by that key.
A secret API key is expected for this integration and is documented, but that key may authorize sensitive device reads and writes depending on its permissions.
`STARLING_API_KEY` | **Yes** | **Yes** | API key created in the Starling Home Hub app
Generate the least-privileged key possible, prefer read-only keys for status checks, rotate the key if exposed, and avoid sharing the environment where the key is set.
A local user or monitoring tool with access to process arguments or logs could potentially see the Starling API key during a request.
The script sends the API key as a URL query parameter, which the Starling API requires and the documentation discusses. Because it is passed to curl as part of the command URL, local process-list or logging exposure may still be possible while requests run.
"${BASE_URL}${path}?key=${API_KEY}"Run this only on trusted machines, avoid shared shells, keep the hub local-only, use HTTPS or certificate pinning where possible, and rotate the key if you suspect local exposure.
Users have less external provenance assurance for a skill that controls sensitive smart-home devices.
The skill is a community integration with unknown source provenance. The included script is visible and there is no remote install step, so this is a provenance note rather than evidence of malicious behavior.
Source: unknown
Review the included script before use and install only if you trust the publisher and understand the device permissions granted by the API key.