Home Assistant Agent (Secure)

ReviewAudited by ClawScan on May 10, 2026.

Overview

This instruction-only skill is transparent about using a restricted Home Assistant token to send smart-home commands through Home Assistant Assist, with no hidden code found.

Install only if you are comfortable giving OpenClaw controlled access to Home Assistant. Create a dedicated non-admin Home Assistant user, limit its device/entity access, store the token carefully, disable trusted-network login bypass if enabled, and remove curl's -k option when your Home Assistant certificate is trusted.

Findings (2)

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 token is over-privileged or leaked, someone could access or control more of Home Assistant than this skill intends.

Why it was flagged

The skill requires a Home Assistant long-lived token, and the artifacts clearly disclose that the token itself is broader than the intended Assist endpoint.

Skill content
The Long-Lived Access Token is not technically scoped to the Conversation API — it grants access to all HA API endpoints available to that user.
Recommendation

Use a dedicated non-admin Home Assistant user with the smallest possible area/entity access, rotate the token if exposed, and avoid storing it where other tools or users can read it.

What this means

The skill can control smart-home devices through Home Assistant, and using -k on an untrusted network could make the token easier to intercept.

Why it was flagged

The skill uses curl to send authenticated commands to Home Assistant; this is central to the purpose, but the -k option disables certificate verification.

Skill content
curl -sk -X POST "$HOME_ASSISTANT_URL/api/conversation/process" ... -H "Authorization: Bearer $HOME_ASSISTANT_TOKEN"
Recommendation

Remove -k when using a trusted certificate, keep Home Assistant on a trusted network, and review commands before allowing high-impact device actions.